U.S. patent application number 11/779216 was filed with the patent office on 2009-01-22 for cochlear implant utilizing multiple-resolution current sources and flexible data encoding.
This patent application is currently assigned to Nurotron Biotechnology, Inc.. Invention is credited to Hongbin Chen, Qian-Jie Fu, Xiaoan Sun, Fan-Gang Zeng.
Application Number | 20090024184 11/779216 |
Document ID | / |
Family ID | 40265464 |
Filed Date | 2009-01-22 |
United States Patent
Application |
20090024184 |
Kind Code |
A1 |
Sun; Xiaoan ; et
al. |
January 22, 2009 |
COCHLEAR IMPLANT UTILIZING MULTIPLE-RESOLUTION CURRENT SOURCES AND
FLEXIBLE DATA ENCODING
Abstract
A programmable cochlear implant system utilizes
multiple-resolution current sources and flexible data-encoding
scheme for transcutaneous transmission. In certain embodiments, the
number of current sources may be equal to or greater than 2, but
equal or less than N-1, where N is the number of electrodes. The
multi-resolution current source may introduce offset currents to
achieve perceptually-based multiple resolutions with high
resolution at low amplitudes and low resolution at high amplitudes.
The flexible data-encoding scheme may allow arbitrary waveforms in
terms of phase polarity, phase duration, pseudo-analog-waveform,
while producing high-rate and high-temporal-precision stimulation.
In one embodiment, a 2-current-source system may support
simultaneous and non-simultaneous stimulation as well as monopolar,
bipolar, pseudo-tripolar, and tripolar electrode
configurations.
Inventors: |
Sun; Xiaoan; (Irvine,
CA) ; Chen; Hongbin; (Irvine, CA) ; Fu;
Qian-Jie; (Arcadia, CA) ; Zeng; Fan-Gang;
(Irvine, CA) |
Correspondence
Address: |
CROWELL & MORING LLP;INTELLECTUAL PROPERTY GROUP
P.O. BOX 14300
WASHINGTON
DC
20044-4300
US
|
Assignee: |
Nurotron Biotechnology,
Inc.
Irvine
CA
|
Family ID: |
40265464 |
Appl. No.: |
11/779216 |
Filed: |
July 17, 2007 |
Current U.S.
Class: |
607/57 ;
607/137 |
Current CPC
Class: |
A61N 1/36038
20170801 |
Class at
Publication: |
607/57 ;
607/137 |
International
Class: |
A61N 1/36 20060101
A61N001/36 |
Claims
1. A method for stimulating a plurality of electrodes implanted in
a human inner ear comprising: multiplexing a plurality of current
sources in accordance with a stimulation mode set using a command
frame; encoding stimulation data for the stimulation of the
plurality of electrodes in a data frame following the command
frame, wherein the stimulation data includes electrode address
information, phase polarity information and amplitude information;
generating a stimulation pulse based on said stimulation data and
using one or more of the plurality of current sources in accordance
with the stimulation mode; and delivering the stimulation pulse to
one or more of the plurality of electrodes in accordance with the
stimulation data.
2. The method of claim 1, wherein encoding comprises encoding
stimulation data for two or more of the plurality of electrodes
into a single data frame.
3. The method of claim 1, wherein the stimulation pulse begins and
ends within a data frame, and wherein the stimulation data in the
data frame further includes pulse width and inter-phase gap
information.
4. The method of claim 3, wherein generating the stimulation pulse
comprises generating the stimulation pulse using one or both of the
pulse width and inter-phase gap information as a clock signal for
timing control.
5. The method of claim 1, wherein the plurality of current sources
includes between 2 and N-1 current sources, where N is the number
of electrodes in said plurality of electrodes.
6. The method of claim 5, wherein the plurality of current sources
consists of 2 current sources.
7. The method of claim 1, wherein the stimulation mode is a mode
selected from the list consisting of: a monopolar stimulation mode,
a bipolar stimulation mode and a pseudo-tripolar stimulation
mode.
8. The method of claim 1, wherein multiplexing comprises
multiplexing the plurality of current sources to achieve
simultaneous and non-simultaneous virtual channels.
9. The method of claim 1, wherein the stimulation pulse comprises
two phases, and wherein the phase polarity information comprises a
polarity for each of the two phases, said polarity being the same
for each of the two phases of the stimulation pulse.
10. The method of claim 9, a second phase of the stimulation pulse
is connected to a first phase of a subsequent pulse to enable
tri-phasic stimulation pulses.
11. The method of claim 1, wherein encoding the stimulation data
comprises encoding the stimulation data in accordance with a
continuous-interleaved-sampling strategy (CIS), wherein the CIS is
selected from the group consisting of: a non-overlapping CIS, a
high-rate CIS, an overlapping CIS and an alternating monophasic
CIS.
12. The method of claim 1, wherein at least one of the plurality of
current sources is configured to produce a multi-resolution current
based on a reference current.
13. The method of claim 1, wherein the plurality of current sources
further comprises a plurality of offset current sources configured
to produce a multi-resolution offset current based on the reference
current.
14. The method of claim 1, wherein delivering the stimulation pulse
comprises transmitting the stimulation pulse transcutaneously to
the plurality of electrodes.
15. A programmable cochlear implant system comprising: a plurality
of electrodes implanted in a human inner ear; a plurality of
current sources multiplexed in accordance with a stimulation mode
set using a command frame; and a processing circuit including a
plurality of current sources and electrically connected to the
plurality of electrodes, the processing circuit configured to:
encode stimulation data for the stimulation of the plurality of
electrodes in a data frame following the command frame, wherein the
stimulation data includes electrode address information, phase
polarity information and amplitude information, generate a
stimulation pulse based on said stimulation data and using one or
more of the plurality of current sources in accordance with the
stimulation mode, and deliver the stimulation pulse to one or more
of the plurality of electrodes in accordance with the stimulation
data.
16. The programmable cochlear implant system of claim 15, wherein
the processing circuit is further configured to encode the
stimulation data for two or more of the plurality of electrodes
into a single data frame.
17. The programmable cochlear implant system of claim 15, wherein
the stimulation pulse begins and ends within a data frame, and
wherein the stimulation data in the data frame further includes
pulse width and inter-phase gap information.
18. The programmable cochlear implant system of claim 17, wherein
the processing circuit is further configured to generate the
stimulation pulse by using one or both of the pulse width and
inter-phase gap information as a clock signal for timing
control.
19. The programmable cochlear implant system of claim 15, wherein
the plurality of current sources includes between 2 and N-1 current
sources, where N is the number of electrodes in said plurality of
electrodes.
20. The programmable cochlear implant system of claim 19, wherein
the plurality of current sources consists of 2 current sources.
21. The programmable cochlear implant system of claim 15, wherein
the stimulation mode is a mode selected from the list consisting
of: a monopolar stimulation mode, a bipolar stimulation mode and a
pseudo-tripolar stimulation mode.
22. The programmable cochlear implant system of claim 15, wherein
the plurality of current sources are multiplexed in accordance with
the stimulation mode to achieve simultaneous and non-simultaneous
virtual channels.
23. The programmable cochlear implant system of claim 15, wherein
the stimulation pulse comprises two phases, and wherein the phase
polarity information comprises a polarity for each of the two
phases, said polarity being the same for each of the two phases of
the stimulation pulse.
24. The programmable cochlear implant system of claim 23, a second
phase of the stimulation pulse is connected to a first phase of a
subsequent pulse to enable tri-phasic stimulation pulses.
25. The programmable cochlear implant system of claim 15, wherein
the processing circuit is further configured to encode the
stimulation data in accordance with a
continuous-interleaved-sampling strategy (CIS), wherein the CIS is
selected from the group consisting of: a non-overlapping CIS, a
high-rate CIS, an overlapping CIS and an alternating monophasic
CIS.
26. The programmable cochlear implant system of claim 15, wherein
at least one of the plurality of current sources is configured to
produce a multi-resolution current based on a reference
current.
27. The programmable cochlear implant system of claim 15, wherein
the plurality of current sources further comprises a plurality of
offset current sources configured to produce a multi-resolution
offset current based on the reference current.
28. The programmable cochlear implant system of claim 15, wherein
the processing circuit is further configured to deliver the
stimulation pulse by transmitting the stimulation pulse
transcutaneously to the plurality of electrodes.
Description
FIELD OF THE INVENTION
[0001] The present invention relates to an implanted auditory
prosthesis that utilizes multiple-resolution current sources and a
data encoding scheme.
BACKGROUND
[0002] Cochlear implants are electronic medical device to help deaf
or severely-hearing-impaired people. They typically consist of an
external signal processor, a transmission coil, an implantable
package with a receiver coil, a hermetically sealed circuit, and an
electrode array. More particularly, these systems include a
microphone to receive sounds and convert them into corresponding
electrical signals. These electrical signals may then be processed
to generate a series of stimulation pulses that are delivered to
the inner ear using a series of implanted electrodes. The
stimulation of these implanted electrodes allows the implantee to
perceive the corresponding ambient sounds.
[0003] A typical cochlear implant includes both an external
component and an internal component. The external component will
typically include a microphone, a speech processor, and a
radio-frequency transmitter, while the internal component includes
an implanted receiver, a hermetically-sealed decoding circuit, and
a series of implanted electrodes. However, there are also numerous
other designs currently available. Regardless of the specific
configuration, a basic premise of all cochlear implant devices is
that ambient sounds are detected by the microphone and a transduced
signal representative of this signal is then generated. The
transduced signal is then processed by a speech processor in
accordance with one of several possible strategies.
[0004] One of the primary design considerations for cochlear
implants is the current source design. To that end, there are two
primary types of current source designs currently in use. The first
is to use one current source for all N electrodes, while the second
approach is to use N current sources for N electrodes. Some
products even use 2N current sources for N electrodes for more
flexibility. Each solution has its own advantages and
disadvantages. For example, with one current source for all
electrodes, the size, complexity, and power consumption of current
stimulator are low. But the stimulation mode is also restricted by
the ability of one current source. No simultaneous stimulation,
current steering, or multi-polar stimulation strategies are
supported. For N (or 2N) current sources for N electrodes, more
flexibility and functionality of stimulation are achieved at the
expense of size, complexity, and power consumption.
[0005] Current resolution is an important factor for current source
design, especially for low stimulation level. For cochlear implant
users, the ratio of current variation versus current .DELTA.l/I is
more important than the current variation .DELTA.I itself.
Traditional linear step-size current source uses a constant
.DELTA.I. Therefore, at low stimulation level where I is small,
.DELTA.I/I is large. To lower .DELTA.I/I, one solution is to
increase the number of current amplitude bit. However, this results
in an increase in the number of current sources in the internal
circuit and lowers the stimulation rate (8-bit requires 256 unit
current sources and 10-bit requires 1024 unit current sources).
When the stimulation level is close to the most comfortable
loudness (MCL) and the value I is large, .DELTA.I/I is usually too
small and the resolution space is wasted. As such, there is a need
for an improved cochlear implant which provides a more balanced
solution for both complexity and functionality.
BRIEF SUMMARY OF THE INVENTION
[0006] Disclosed and claimed herein is a programmable cochlear
implant that utilizes multiple current sources and a flexible
data-encoding scheme. In one embodiment of the invention, a method
for stimulating electrodes implanted in a human inner ear includes
the acts of multiplexing current sources in accordance with a
stimulation mode set using a command frame, and encoding
stimulation data for the stimulation of the plurality of electrodes
in a data frame following the command frame, where the stimulation
data include electrode address information, phase polarity
information and amplitude information. The method further includes
generating a stimulation pulse based on the stimulation data and
using one or more of the current sources in accordance with the
stimulation mode, and delivering the stimulation pulse to the
electrodes in accordance with the stimulation data.
[0007] Other aspects, features, and techniques of the invention
will be apparent to one skilled in the relevant art in view of the
following detailed description of the invention.
BRIEF DESCRIPTION OF THE DRAWINGS
[0008] FIG. 1 is a block diagram of one embodiment of a cochlear
prosthesis in which two current sources are utilized;
[0009] FIG. 2A is the data encoding format supporting from two
current sources in accordance with one embodiment;
[0010] FIG. 2B is one embodiment of a scheme of self-timing with
pulse-width-modulation bit coding;
[0011] FIG. 3 depicts stimulation modes supported by the data
encoding scheme of one embodiment of the invention;
[0012] FIGS. 4A-4C depict embodiments of switch networks for
various stimulation modes;
[0013] FIGS. 5A-5F depict embodiments of command and data frames
for various stimulation modes in accordance with the principles of
the invention;
[0014] FIGS. 6A-6B depict an arbitrary waveform generator according
to one embodiment of the invention;
[0015] FIG. 7 illustrates one embodiment of a nonlinear step size
current source implementation scheme;
[0016] FIG. 8 shows one embodiment of a schematic for a reference
current selection and offset current control circuit; and
[0017] FIGS. 9A-9C illustrate various
continuous-interleaved-sampling strategy (CIS) implementations in
accordance with one or more embodiments of the invention.
DETAILED DESCRIPTION OF EXEMPLARY EMBODIMENTS
[0018] Described and claimed herein is a programmable cochlear
implant system utilizing multiple-resolution current sources and a
flexible data-encoding scheme. In one embodiment, the system can
support simultaneous and non-simultaneous stimulation as well as
monopolar, bipolar, pseudo-tripolar, and tripolar electrode
configurations.
[0019] One aspect of the invention is a cochlear implant having 2
to N-1 current sources for N electrodes. In one embodiment, the
number of amplitude bits are balanced with the current resolution
at a low stimulation level. By introducing an offset current
circuit consisting of three current sources whose states are
controlled by 2-bit range information, a four- range current source
may be implemented with four different current resolutions of I,
2I, 4I, and 8I, where I is the minimum reference current. Compared
with traditional single range current source, this embodiment
improves current resolution by 4 times at low stimulation level,
according to one embodiment. In another embodiment, the invention
provides high resolution by not overlapping resolution space among
different ranges.
[0020] Another aspect of the invention is a highly-flexible data
encoding scheme to support the aforementioned design of 2 to N-1
current sources. The information for each current source may be
modularized and conveniently added or removed from a data frame.
With each module, the phase polarity, amplitude, stimulating
electrode, and phase duration of electrical stimulation can be
individually set for each current source. Therefore, flexible
stimulation modes, stimulation strategies, arbitrary pulse
polarity, arbitrary stimulation waveforms, flexible pulse duration
and inter-phase gap may all be supported. In one embodiment, the
range of pulse widths may be from 1 .mu.s to 1024 .mu.s. Similarly,
the range of inter-phase gap may be from 0 .mu.s to 31 .mu.s. In
one embodiment, the resolution for both may be essentially 1 .mu.s.
Finally, a high-rate stimulation mode is supported by a special
command specifying only one pulse duration, producing a 31-kHz
overall stimulation rate with 1 current source and a 62-kHz rate
with 2 current sources.
[0021] The timing control for implanted circuit may be an important
factor for the normal operation of a current stimulator. While one
solution has been to add a local timer in the implanted circuit,
this tends to increase the power consumption and lower the
reliability of the circuit. Moreover, any minor defect of the timer
will cause unpredictable error in the internal circuit and current
stimulator. Also, it is very difficult to synchronize the timing
between external timer and internal timer. To achieve a reliable
synchronization between the two timers, usually a phase lock loop
(PLL) circuit is required in the implanted circuit, which itself is
complicated and power hungry. Thus, another aspect of the invention
is to provide timing from outside such that no timer or PLL circuit
is required inside. As will be described in more detail below, with
a careful design of bit coding which enables signal level changes
at the beginning of each data bit, this data encoding scheme can
provide timing information to the implanted circuit, in addition to
providing power and data.
[0022] Referring first to FIG. 1, depicted is one embodiment of a
cochlear prosthesis 100 configured to implement one or more
embodiments of the invention. As shown, the cochlear prosthesis 100
includes a cool/transformer 105 coupled to a rectifier/LPF power
supply 110 and a processing circuit 115. In one embodiment, the
power supply 110 is itself electrically coupled to the processing
circuit 115, as shown in FIG. 1.
[0023] While in the embodiment of FIG. 1 processing circuit 115
includes two current sources 120.sub.1 and 120.sub.2, it should
equally be appreciated that additional current sources may
similarly be included. As shown, current sources 120.sub.1 and
120.sub.2, are each coupled to multiplexers 125.sub.1 and
125.sub.2, respectively. In turn, multiplexers 125.sub.1 and
125.sub.2 may provide current signals to a plurality of electrodes
130, in accordance with a selected stimulation mode. While 24
electrodes are depicted in FIG. 1, it should equally be appreciated
that more or fewer electrodes may be included with the cochlear
prosthesis 100 of FIG. 1.
[0024] Processing circuit 115 is further depicted as including a
data decoder 135 for decoding the incoming data, and mode detector
140 for detecting the mode of the incoming data. Once the data is
decoded and the mode detected, the data distributor 145 may use
this information to control the current sources 120.sub.1 and
120.sub.2, timing control 150 and electrode selector 155, as shown
in FIG. 1. As shown, the timing control 150 and electrode selector
may be used to selectively stimulate one or more of the plurality
of electrodes 130 using one or both of the current sources
120.sub.1 and 120.sub.2. The processing circuit 115 may further
comprise a voltage sampler 160 and backward data coder 165 for
transmitting information regarding electrode impedance, electrical
field potentials, current flow through the internal receiving coil,
data decoding status, and evoked neural activities. It should be
understood that, with the exception of the plurality of electrodes
130, the other components of the cochlear prosthesis 100 may
comprise an external portion, meaning that such components may not
be implanted under the skin or residing within the inner ear.
[0025] With the 2-current-source configuration of FIG. 1 and a
highly-flexible data encoding scheme as described herein, a large
variety of stimulation modes and strategies may be implemented. If
only one current source were used, only standard mono-polar or
bipolar CIS stimulation pulses can be generated. By using two
current sources simultaneously (e.g., current sources 120.sub.1 and
120.sub.2), traditional non-overlapping CIS, as well as overlapping
CIS (virtual channel) and even alternating monophasic CIS are
enabled.
[0026] Referring now to FIG. 2A, depicted is one embodiment of a
data frame 200 comprised of a total of 50 bits. In the depicted
embodiment, the individual bits include:
TABLE-US-00001 Bits 1-2: 2 bits, start of a data frame. Bits 3-7: 5
bits electrode info for current source 1, named Pulse1. Bit 8: 1
bit sign info of first phase of Pulse1, "0" negative, "1" positive.
Bit 9: 1 bit sign info of second phase of Pulse1 "0" negative, "1"
positive. Bits 10-17: 8 bits amplitude info of Pulse1. Bit 18: 1
bit parity check for Pulse1 info, bits 3-17. Bits 19-23: 5 bits
electrode info for next pulse of current source 2, named Pulse2.
Bit 24: 1 bit sign info of first phase of Pulse2, "0" negative, "1"
positive. Bit 25: 1 bit sign info of second phase of Pulse2, "0"
negative, "1" positive. Bits 26-33: 8 bits amplitude info of
Pulse2. Bit 34: 1 bit parity check for Pulse2 info, bits 19-33.
Bits 35-44: 10 bits phase width. Bits 45-49: 5 bits inter-phase
gap. Bit 50: 1 bit parity check for phase info, bits 35-49.
[0027] The bit coding of the proposed data encoding scheme may be
used to provide timing control for internal pulse generation. In
this way, the implanted circuit may not require a local timer.
After electrode and amplitude information are provided in the
beginning of a data frame, the remaining bits in a data frame
(e.g., data frame 200) provide pulse width and inter-phase gap
information and may also act as clock signal for the timing control
of current pulse. Phase extending bits can be added after a data
frame to generate long phase duration pulses. The start and end of
each phase of a pulse may be synchronized by the onset of bits in a
data frame. In one embodiment, each bit may include of 10 to 15 RF
cycles to provide cycle error tolerance. It should be appreciated
that the proposed timing control may make the implanted circuit
more reliable and easier to implement.
[0028] As previously mentioned, one embodiment of the coding scheme
may support flexible 2 to N-1 current sources for N electrodes. For
each current source, one embodiment of the coding scheme has 16
bits in date frame 200 corresponding to it, including 5-bit
electrode address information, 2-bit phase polarity information and
8-bit pulse amplitude information, and 1 parity check bit. In this
fashion, multiplexing of 2 or more current sources to achieve
monopolar, bipolar, and pseudo-tripolar stimulation modes is
enabled. The pseudo-tripolar refers to apically or basally applied
negative current to sharpen the electric field. A true tripolar
will sharpen the field from both sides. In certain embodiments,
alternating mono-phasic stimulus may result in the power
consumption being at least as good as regular bipolar stimulation.
In addition, with 3 or more current sources true tripolar
stimulation may be provided.
[0029] It should further be appreciated that multiplexing 2 or more
current sources may achieve both simultaneous and non-simultaneous
virtual channels. To that end, in one embodiment the total number
of virtual channel may be at least N+(N-1), where N is the number
of electrodes.
[0030] In certain embodiments, a pulse (e.g., Pulse1, Pulse2)
always starts and finishes within one data frame. After electrode,
amplitude and phase polarity information are provided, the width
and phase gap bits in a data frame may act as a clock signal for
the timing control of present pulse. The start and end of each
phase of a pulse may be synchronized by the onset of bits in a data
frame.
[0031] Referring now to FIG. 2B, depicted is one embodiment of a
bit coding scheme 210 in which pulse width modulation with 15 radio
frequency (RF) cycles are used. In one embodiment a "0" bit has 5
on-cycles and 10 off-cycles. Similarly, a "1" bit may have 10
on-cycles and 5 off-cycles. Redundant cycles may be included to
tolerate up to 2-cycle errors. By way of example, Table 1 below
shows one embodiment of a decoding scheme:
TABLE-US-00002 TABLE 1 Decoding Scheme Error on-cycle .ltoreq.2 "0"
3 .ltoreq. on-cycle .ltoreq. 7 "1" 8 .ltoreq. on-cycle .ltoreq. 12
Error on-cycle .gtoreq.13
[0032] In one embodiment, each bit must start with an on-cycle and
end with an off-cycle. In this way, the start of a bit may be
associated with a rising edge, which can be used as a clock signal
to trigger other events in the implanted circuit.
[0033] In one embodiment, the flexible coding scheme of the
invention may provide a high temporal resolution, with the shortest
pulse duration being set to 8 .mu.s and a temporal resolution set
to one period of data bit (0.5 .mu.s). This high temporal
resolution may also allow accurate encoding of fundamental
frequency (F0) and frequency modulated (FM) information.
Stimulation Mode /
[0034] Referring now to FIG. 3 depicted are at least some of the
stimulation modes supported by the data encoding scheme of one
embodiment of the invention. In particular, an internal electrode
configuration 310 for a monopolar stimulation mode is shown. In
addition, FIG. 3 further depicts an internal electrode
configuration 320 for a bipolar stimulation mode, an internal
electrode configuration 330 for a pseudo-tripolar stimulation mode,
and an internal electrode configuration 340 for a tripolar
stimulation mode. In this fashion, one embodiment of the invention
may provide a flexible encoding scheme for a variety of stimulation
modes.
[0035] Referring now to FIGS. 4A-4C, depicted are exemplary switch
networks for various stimulation modes, in accordance with the
principles of the invention. As shown, a flexible coding scheme of
the invention may allow high power and coding efficiency, while
providing arbitrary waveform outputs, including alternating or
consecutive monophasic pulses for power and electric stimulation
efficiency. To that end, FIG. 4A depicts a switching network 400
comprising of a first current source 410 and a second current
source 420, electrically connected to a plurality of electrodes
430.sub.1-430.sub.n. As will be understood by one in the art, when
one or more of the switches of switch network 400 are closed, a
voltage VDD may be provided to stimulate one or more of the
plurality of electrodes 430.sub.1-430.sub.n.
[0036] FIG. 4B depicts the switching network 400 in which bipolar
stimulation of the plurality of electrodes 430.sub.1-430.sub.n is
being implemented using the first current source 410 and the second
current source 420. As shown, switches 435, 440, 445 and 450 have
been closed in order to implement bipolar stimulation. It should of
course be understood that numerous other switching arrangements may
be used in accordance with the invention.
[0037] Referring now to FIG. 4C, depicted is the switching network
400 implementing tripolar stimulation of the plurality of
electrodes 430.sub.1-430.sub.n using the first current source 410
and the second current source 420. As shown, in addition to having
switches 435 and 440 closed, switches 455 and 460 are also closed.
In addition, switches 445 and 450 have been opened in order to
implement the tripolar stimulation scheme. As with the embodiment
of FIG. 4B, it should of course be understood that numerous other
switching arrangements may be used in accordance with the
invention.
[0038] In certain embodiments, the stimulation mode may be set in
the command frame not in the data frame. As previously mentioned,
four of the possible stimulation modes include bipolar, monopolar,
pseudo-tripolar and tripolar.
[0039] Bipolar stimulation can be implemented by either using one
current source multiplexing between different electrodes, or by
using two current sources in a monopolar mode. By way of example,
FIG. 5A depicts one embodiment of a command frame 500 for the
bipolar mode at a first electrode (e.g., one of electrodes 130)
using a first current source (e.g., current source 120.sub.1).
Similarly, an exemplary data frame 510 is also depicted in FIG.
5A.
[0040] By way of providing another example of bipolar stimulation,
FIG. 5B depicts another exemplary command frame 520 and
corresponding data frame 530 for a first electrode, but in this
case using two current sources (e.g., current source 120.sub.1 and
current source 120.sub.2). In one embodiment, the polarity of the
two current sources may be opposite in order to cancel out the
electrode field outside of the two electrodes.
[0041] With respect to a monopolar stimulation mode, one embodiment
of a command frame 540 and data frame 550 using one current source
are depicted in FIG. 5C.
[0042] In the case of a Pseudo-tripolar stimulation mode, one
embodiment of a command frame 560 and data frame 570 using one
current source are depicted in FIG. 5D. In the pseudo-tripolar mode
of FIG. 5D, only one current source may be used and the returning
two electrodes may be grounded.
[0043] An exemplary command frame 580 and data frame 590 for a
tripolar stimulation mode are shown in FIG. 5E. As shown, two
current sources are used, and each returning electrode drain is
half of the total current.
[0044] Finally, a special high-rate stimulation mode can be
achieved using a high-rate mode command using a high rate data
frame 595, as shown in the embodiment of FIG. 5F. Information
regarding stimulation mode, pulse duration, gap duration, and
current source may be transmitted before stimulation. In certain
embodiments, only information regarding electrode, pulse polarity,
and amplitude may be updated for cycle. Stimulation may start as
soon as all 18 bits are transferred and decoded. The overall
stimulation rate can be as high as 31-kHz with 1 current source and
62-kHz with 2 current sources.
Arbitrary Waveform Generation
[0045] Referring now to FIGS. 6A-6B, depicted is an arbitrary
waveform 600 generated using a flexible phase polarity, in
accordance with one embodiment of the invention. As shown, each
pulse comprises two phases thereby forming a bi-phasic period. The
total charge for one period of the sinusoid 610 may be zero,
meaning that the total charge is balanced. However, unlike the
typical case, the polarity of each of the two phases of a pulse,
such as pulse 620, may be arbitrarily assigned. This may allow for
alternating phase pulses (negative-positive, positive-negative) and
monolithic phase pulses (negative-negative, positive-positive),
unlike prior art embodiments. This feature may be especially useful
to generate pseudo-analog-waveform stimulations, where the biphasic
pulses are not charge balanced for each individual pulse, but the
accumulated long term charge is balanced over the entire sinusoid
610, as shown in FIG. 6A. In one embodiment, this is enabled using
the previously-described two polarity bits for the two individual
phases. While it should be appreciated that the sinusoid 610 may be
longer or shorter, in one embodiment the sinusoid 610 may be
approximately 1000 .mu.s, while the bi-phasic period 630 may be
approximately 50 .mu.s. Obviously, different values may assigned to
each of the sinusoid 610 and/or period 630.
[0046] Referring now to FIG. 6B, depicted is an enlarged view of
the bi-phasic period 620 of FIG. 6A. As shown, period 620 is
comprised of a first phase 640 and a second phase 650 each having
the same polarity, and thus the same charge over the entire period
630. The next period can be programmed to have two pulses with the
same amplitude but the opposite polarity. In certain embodiments,
this flexible coding may be used to produce mono-phasic or
tri-phasic waveforms, which have the advantage of lower stimulation
thresholds and possibly more focused electrical fields than
biphasic waveform. Longer battery life and possibly better
performance can be achieved.
Strategy Implementation
[0047] In the practice of programmable current source design, there
has been two general approaches. The first approach, which is now
largely obsolete, is to control the gate-source voltage to get
variable drain-source current. This method, used by first
generation cochlear implant products, required that each current
source be calibrated due to the fact that the nonlinear
V.sub.GS-I.sub.DS relationship has a large variation among
transistors.
[0048] The second approach for current source design is to use a
linear combination of fixed value current sources to get a desired
current value. Usually, a group of high precision fixed value
currents is used as reference currents to generate output current
values by current mirroring. The majority of current cochlear
implant products use this type of current source.
[0049] For this type of current sources, to achieve a higher
stimulating accuracy, a smaller step size .DELTA.I of current
increment is required, under a given current range [I.sub.min,
I.sub.max]. Thus, more current amplitude bits B are needed to get a
smaller step size, as shown below:
.DELTA. I = I max - I min 2 B ##EQU00001##
[0050] However, more amplitude bits B usually means more unit
current sources. For example, B=8 requires 255 unit current
sources, and B=10 requires 1023 unit current sources. For an
integrated circuit implementation, it is undesirable to have so
many current sources, since more chip space is required causing a
parasitic effect.
Current Source Implementation
[0051] Consider the example of B=8, I.sub.min=0, I.sub.max=2 mA,
where there are 256 current levels with a step size of 8 .mu.A. For
small stimulations near threshold level T, this step size may be
too large such that the actual T level might fall between two
current levels. For large stimulations near the MCL, this step size
may be too small such that different current levels make no
difference to patients, thus wasting limited current levels. Thus,
it may be desirable to use a small step size for small currents to
get accurate T levels, and a large step size for large currents for
adequate sensational variation.
[0052] One embodiment of a current source control scheme 700 in
accordance with the principles of the invention is depicted in FIG.
7. In one embodiment, the current control scheme 700 supports
nonlinear step-sizes for different current resolutions
710.sub.1-710.sub.n, of I, 2I, 4I, 8I, 16I and 32I, for example,
where I is the minimum reference current. In one embodiment, the
amplitude may be encoded with a 6-bit value and the range with a
2-bit value, as described above with reference to FIG. 2. In one
embodiment, an offset current circuit may be introduced. By way of
a non-limiting example, the scheme 700 includes current sources
720.sub.1-720.sub.n usable to provide an offset current, and whose
states may be controlled by the 2-bit range information.
[0053] Referring now to FIG. 8, depicted is one embodiment of a
control circuit 800 for reference current and offset current
selection. In particular, reference current selector 810 decodes
the 2-bit range to select one of four possible reference currents,
(i.e., 2 .mu.A, 4 .mu.A, 8 .mu.A and 16 .mu.A). This information
may then be used to control the offset current source 820, which in
one embodiment is based on the selection of current sources
720.sub.1-720.sub.n from FIG. 7. In addition, the current selector
810 may further be configured to use the 6-bit amplitude
information to control a 6-bit current digital-to-analog converter
(DAC) 830. In one embodiment, the 6-bit DAC and the offset current
source may use the same reference current I, in which case only one
reference current generator may be needed. The following Table 2
depicts the exemplar bit controlled values for the control circuit
800:
TABLE-US-00003 TABLE 2 Exemplary Bit Controlled Current Values
Range Reference Minimum Bits Current Offset Current Current Maximum
Current 00 2 .mu.A (0 + 0 + 0) * 2 = 0 .mu.A 0 .mu.A 0 + (2.sup.6 -
1) * 2 = 126 .mu.A 01 4 .mu.A (32 + 0 + 0) * 4 = 128 .mu.A 128
.mu.A 128 + (2.sup.6 - 1) * 4 = 380 .mu.A 10 8 .mu.A (32 + 16 + 0)
* 8 = 384 .mu.A 384 .mu.A 384 + (2.sup.6 - 1) * 8 = 888 .mu.A 11 16
.mu.A (32 + 16 + 8) * 16 = 896 .mu.A 896 .mu.A 896 + (2.sup.6 - 1)
* 16 = 1904 .mu.A
[0054] Table 2 above illustrates that for relatively small currents
(e.g., 0-126 .mu.A) the step size is 2 .mu.A. For larger currents
(896 .mu.A-1904 .mu.A), the step size is shown as being 16 .mu.A.
In the depicted embodiment, the current source uses 8 amplitude
bits and a 6-bit DAC to achieve a minimal step size of a 10-bit
DAC. Compared with 8-bit DAC, the 6-bit DAC uses 53% less unit
current sources, and when compared with a 10-bit DAC, the 6-bit DAC
uses 88% less unit current sources.
[0055] In certain embodiments, 8-bit 256 level nonlinear step size
amplitude control may provide one or more of the following
features: [0056] Current precision improved: from 8 .mu.A to 2
.mu.A. [0057] Same 8 amplitude bits, rather than 10 bits. [0058]
Uses 6-bit DAC rather than 10-bit DAC. [0059] No command frame
needed to set range. [0060] Total number of unit current sources
decreases from 255 (8-bit DAC) or 1023 (10-bit DAC) to 119.
[0061] FIGS. 9A-9C depict various embodiments of CIS strategies in
accordance with certain embodiments of the invention. In FIG. 9A,
for example, a monopolar non-overlapping CIS strategy 900 is
depicted. As shown, following command frame 905, data frames 910
are received for each of 24 possible channels for a two current
source embodiment. In one embodiment, the data frame 910 may be
configured in accordance with the embodiment of FIG. 2. In
addition, each of the 24 possible channels may correspond to an
individual electrode to be stimulated (e.g., electrodes 130 of FIG.
1).
[0062] Continuing to refer to FIG. 9A, the depicted Chn1 data frame
is decoded to produce the corresponding pulse for Chn1 after delay
915 using either the first current source (cs1) or the second
current source (cs2).
[0063] Similarly, the Chn2 data frame is decoded to produce the
corresponding pulse for Chn2, as shown in FIG. 9A. It should be
noted that the pulses for each of Chn1-Chn24 do not overlap since
FIG. 9A corresponds to one embodiment of a monopolar
non-overlapping CIS strategy. Moreover, each pulse of FIG. 9A
further includes an amplitude encoded into the corresponding data
frame, as described above, and may be generated using either
current source.
[0064] FIG. 9B depicts one embodiment of an overlapping CIS
strategy 925. As with the embodiment of FIG, 9A, a command frame
930 precedes data frames 935 for each of 24 possible channels 940,
where each channel may correspond to an individual electrode to be
stimulated (e.g., electrodes 130 of FIG. 1). However, unlike the
non-overlapping CIS strategy 900, each data frames 935 of the
overlapping CIS strategy 935 may include stimulation data for two
channels - e.g., Chn1 and Chn13, Chn2 and Chn14, Chn3 and Chn15,
etc. Thus, the depicted Chn1,13 data frame may be decoded to
produce the pulse for Chn1 using cs1, as well as the pulse for
Chn13 using cs2. In addition, these pulses may overlap, as shown in
FIG. 9B. As with the embodiment of FIG. 9A, the embodiment of 9B
includes delay 945 which is a result of the decoding process.
[0065] FIG. 9C illustrates another embodiment of a CIS strategy 950
in which two current sources are multiplexed to achieve
simultaneous and non-simultaneous virtual channels. In one
embodiment, the total number of virtual channels can be at least
N+(N-1), where N equals the number of electrodes. In another
embodiment, more virtual channels are possible if the amplitudes
between two adjacent channels are manipulated.
[0066] While the invention has been described in connection with
various embodiments, it should be understood that the invention is
capable of further modifications. This application is intended to
cover any variations, uses or adaptation of the invention
following, in general, the principles of the invention, and
including such departures from the present disclosure as come
within the known and customary practice within the art to which the
invention pertains.
* * * * *