Re: 120 channels-my impression

[Guest guest]

Hi Virg !

Absolutely - 120 channels are stimulated, but not simultaneously.

Net effect is perceived by the brain as 120 distinct channels.

Others talk about number of electrodes as if this is the same as the

number of channels. NOT SO !! Truth is, by varying how these

electrodes are fired, the number of channels can be much more than

the number of electrodes.

It is the design of the implant that determines how the electrodes

can be fired, and AB is the only one capable of doing the 120

channels today, although I believe that MedEl MAY be able to do so in

the future.

p.s. - this has already been discussed on this forum, but it keeps

being brought up again with the same misconceptions being put out.

Regards,

Jim S.

>

> I got the idea from the interview with with Jo Osberger Ph.D,

> Director of Auditory Clinical Research Advanced Bionics

> that 120 separate areas will be stimulated. From the interview

(excerpt

> below) it seemed that the unequal currents in each pair of adjacent

> electrodes would stimulate an in-between area, depending upon the

ratio of

> currents. Areas not actually contacted directly by any of the

contacts.

>

> Of course, these areas are stimulated in sequence, not

simultaneusly, as

> the 15 or 22 electrodes are not now stimulated simultaneously.

>

> Virg

>

> Excerpt from the interview:

>

> " Current steering allows us to create synthetic or virtual channels

> between electrode contacts to deliver more frequency information to

the

> cochlear-implant user. Prior to this technology, the number of

frequency

> channels was limited by the number of electrode contacts in the

array.

> Through simultaneous delivery of current to pairs of adjacent

electrodes,

> stimulation can be " steered " to sites between the contacts by

varying the

> proportion of current delivered to each electrode, thereby

eliciting pitch

> percepts that are intermediate to the two electrodes. The number of

> distinct pitches that can be heard defines the number of spectral

channels

> that can be perceived by the cochlear implant user. It is predicted

that

> increased spectral resolution will improve music perception and

enjoyment

> in cochlear implant users. "

>

[Guest guest]

Seriously Virg,

I do not understand the excerpt--I got completely lost at what she is

saying. Okay let me see if I understand how it works because I know I

probably will not get it right. But I'll keep trying hahha.

Currently AB has 16 channels as there are only 16 electrode contacts

in the array. So if we have 16 electrodes all firing at the same time

in paired sequence, with current steering (what the heck is current

steering--what does steering mean anywayz?) I would assume that each

electrode has series of sections/areas being stimulated, thus creating

multiple virtual channels which give us more pitch perceptions? Is

this what ms.Osberger is saying below?

>

> I got the idea from the interview with with Jo Osberger Ph.D,

> Director of Auditory Clinical Research Advanced Bionics

> that 120 separate areas will be stimulated. From the interview (excerpt

> below) it seemed that the unequal currents in each pair of adjacent

> electrodes would stimulate an in-between area, depending upon the

ratio of

> currents. Areas not actually contacted directly by any of the contacts.

>

> Of course, these areas are stimulated in sequence, not simultaneusly, as

> the 15 or 22 electrodes are not now stimulated simultaneously.

>

> Virg

>

> Excerpt from the interview:

>

> " Current steering allows us to create synthetic or virtual channels

> between electrode contacts to deliver more frequency information to the

> cochlear-implant user. Prior to this technology, the number of frequency

> channels was limited by the number of electrode contacts in the array.

> Through simultaneous delivery of current to pairs of adjacent

electrodes,

> stimulation can be " steered " to sites between the contacts by

varying the

> proportion of current delivered to each electrode, thereby eliciting

pitch

> percepts that are intermediate to the two electrodes. The number of

> distinct pitches that can be heard defines the number of spectral

channels

> that can be perceived by the cochlear implant user. It is predicted that

> increased spectral resolution will improve music perception and

enjoyment

> in cochlear implant users. "

>

[Guest guest]

I think (although guess would be a better word) that channel steering works

a little like this - the elctrode spacing, although very close, is probably

" miles " apart in terms of cochlear nerve endings, ie two adjacent electrodes

will not be covering two adjacent nerve endings - there will be lots in

between. By firing two adjacent electrodes but with different amounts of

current " bends " the current one way or another and stimulates a nerve ending

other than the one directly below the electrode. The degree of control the

processor has determines the number of " virtual " channels.

Or something like that.

Maybe.

NZ

> Re: 120 channels-my impression

>

>

> Seriously Virg,

>

> I do not understand the excerpt--I got completely lost at

> what she is saying. Okay let me see if I understand how it

> works because I know I probably will not get it right. But

> I'll keep trying hahha.

>

> Currently AB has 16 channels as there are only 16 electrode

> contacts in the array. So if we have 16 electrodes all

> firing at the same time in paired sequence, with current

> steering (what the heck is current steering--what does

> steering mean anywayz?) I would assume that each electrode

> has series of sections/areas being stimulated, thus creating

> multiple virtual channels which give us more pitch

> perceptions? Is this what ms.Osberger is saying below?

>

[Guest guest]

Hi ,

Interesting topic, and I think I am beginning to understand some of

this, I have been trying to decypher all of the geek speak I have

seen to date and I believe your explaination make perfect sense. I

know most of the stuff I have read from the mfg so far is over my

head. I am going to a Hard of Hearing Convention next weekend and AB

is presenting a session on this subject, you can bet I will be front

and center, maybe they will have Pictures and I can understand that

lol.

Thanks for that explaination, it makes perfect sense to me and I bet

it is a darn good guess.

Regards,

Mike " Ears Hopin " P

>

> I think (although guess would be a better word) that channel

steering works

> a little like this - the elctrode spacing, although very close, is

probably

> " miles " apart in terms of cochlear nerve endings, ie two adjacent

electrodes

> will not be covering two adjacent nerve endings - there will be

lots in

> between. By firing two adjacent electrodes but with different

amounts of

> current " bends " the current one way or another and stimulates a

nerve ending

> other than the one directly below the electrode. The degree of

control the

> processor has determines the number of " virtual " channels.

>

> Or something like that.

>

> Maybe.

>

>

> NZ

>

> > Re: 120 channels-my impression

> >

> >

> > Seriously Virg,

> >

> > I do not understand the excerpt--I got completely lost at

> > what she is saying. Okay let me see if I understand how it

> > works because I know I probably will not get it right. But

> > I'll keep trying hahha.

> >

> > Currently AB has 16 channels as there are only 16 electrode

> > contacts in the array. So if we have 16 electrodes all

> > firing at the same time in paired sequence, with current

> > steering (what the heck is current steering--what does

> > steering mean anywayz?) I would assume that each electrode

> > has series of sections/areas being stimulated, thus creating

> > multiple virtual channels which give us more pitch

> > perceptions? Is this what ms.Osberger is saying below?

> >

>

[Guest guest]

We don't currently have 16 electrodes firing at the same time, Nora.

They fire in sequence, I believe.

said it about the way that I understand it. Lots of nerve endings

between adjacent electrodes and, somehow, running different currents in the

electrodes stimulates one (or more?) of the in-between nerve endings.

Different combinations of current stimulate different nerve endings. It

would seem (guessing) that with 8 pairs of adjacent electrodes, 15

different combinations of current for each pair would give access to 120

different nerve endings or groups of nerve endings. I doubt that a single

nerve ending can be targeted.

I think it would be fascinating to read an in depth description of the

process but I doubt that AB will be doing that for awhile yet.

Virg

From: " lil_nora77 " <lil_nora77@...>

Seriously Virg,

I do not understand the excerpt--I got completely lost at what she is

saying. Okay let me see if I understand how it works because I know I

probably will not get it right. But I'll keep trying hahha.

Currently AB has 16 channels as there are only 16 electrode contacts

in the array. So if we have 16 electrodes all firing at the same time

in paired sequence, with current steering (what the heck is current

steering--what does steering mean anywayz?) I would assume that each

electrode has series of sections/areas being stimulated, thus creating

multiple virtual channels which give us more pitch perceptions? Is

this what ms.Osberger is saying below?

[Guest guest]

How does the audiologist doing the mapping decide which of the 120 channels

to use for the 16 electrodes out of interest?

Severe/profoundly deaf since birth

Lost residual hearing October 2005

Approved for CI March 2006

Surgery 9th June 2006

Activation 29th June 2006

_____

From: [mailto: ] On Behalf Of

Virg

Sent: Thursday, 25 May 2006 1:38 PM

Subject: Re: Re: 120 channels-my impression

We don't currently have 16 electrodes firing at the same time, Nora.

They fire in sequence, I believe.

said it about the way that I understand it. Lots of nerve endings

between adjacent electrodes and, somehow, running different currents in the

electrodes stimulates one (or more?) of the in-between nerve endings.

Different combinations of current stimulate different nerve endings. It

would seem (guessing) that with 8 pairs of adjacent electrodes, 15

different combinations of current for each pair would give access to 120

different nerve endings or groups of nerve endings. I doubt that a single

nerve ending can be targeted.

I think it would be fascinating to read an in depth description of the

process but I doubt that AB will be doing that for awhile yet.

Virg

From: " lil_nora77 " <lil_nora77@...>

Seriously Virg,

I do not understand the excerpt--I got completely lost at what she is

saying. Okay let me see if I understand how it works because I know I

probably will not get it right. But I'll keep trying hahha.

Currently AB has 16 channels as there are only 16 electrode contacts

in the array. So if we have 16 electrodes all firing at the same time

in paired sequence, with current steering (what the heck is current

steering--what does steering mean anywayz?) I would assume that each

electrode has series of sections/areas being stimulated, thus creating

multiple virtual channels which give us more pitch perceptions? Is

this what ms.Osberger is saying below?

[Guest guest]

Hi

I'm not real sure but I am sure AB is training the audiologist on

just how to do this. I think there was something along this in

Chorost book Rebuilt; it seems like he touched on this. His

website is www.michaelchorost.com and his book is a great read!!

Kim B

Advanced Bionics CI

>

> How does the audiologist doing the mapping decide which of the 120

channels

> to use for the 16 electrodes out of interest?

>

>

>

>

>

>

>

> Severe/profoundly deaf since birth

>

> Lost residual hearing October 2005

>

> Approved for CI March 2006

>

> Surgery 9th June 2006

>

> Activation 29th June 2006

>

> _____

[Guest guest]

> How does the audiologist doing the mapping decide which of the 120

channels

> to use for the 16 electrodes out of interest?

I can only guess, .

If the electrodes are paired, as was said, and the frequency span overall

is 250Hz to 8Khz, it may be something like this:

7,750 Hz spread/ 8 gives 969 Hz spread per pair. If there are 120

channels, that would amount to 15 frequency bands per pair, each spanning

1/15 of its 969Hz allotment or 66Hz per band. So it seems that possibly

the lower 66hz band (say 250 hz to 312 hz) would be assigned to one of the

lower 16 current combinations, the next 66 hz, to another combination, etc.

for 15 current combinations for that lower pair. And then the next pair of

contacts would span the next 969Hz, splitting it into 15 channels, etc.

All pure conjecture on my part of course. It may be nothing like that.

Virg

[Guest guest]

Virg,

I think your guess seems more accurate than anything else that I

have researched !! Using HiRes P can provide approximately 5,100 pps

on each paired of electrodes, HiRes S can provide approximately

around 2,500 pps on each 16 electrodes. HiRes S and P only provide

CONSTANT (not variable) high stimulation rates on AB implant

wearers. What most of us think that adding 120 or 5 or 32 or minus 2

or 10 does not make any difference which is CORRECT ! It is fact

that I tried that way by substracting and adding electrodes in my

implant still not change my speech or music perceptions.

However; changing the strategy to CIS from HiRes S or P does make my

hearing difference by actually changing the stim rates.

Your explanation by steering the physical current to " in between "

the two adjacent electrodes does even change the rates or Hz.

Virtual channels can narrow the current flow into smaller rates out

of ONE electrode rates like a fraction. Using one half, one tenth,

four seventh of a " whole " electrode firing summations is like

smaller pipes being diverted by one large pipe.

Smaller rates provide lower and middle frequency sounds. 1/6 or

other fraction of one major pipe electrode pps or rates is count as

one channel.

Thats pretty strange !!! Physic talks ...

Rick F

Virg wrote :

" I can only guess, .

If the electrodes are paired, as was said, and the frequency span

overall

is 250Hz to 8Khz, it may be something like this:

7,750 Hz spread/ 8 gives 969 Hz spread per pair. If there are 120

channels, that would amount to 15 frequency bands per pair, each

spanning

1/15 of its 969Hz allotment or 66Hz per band. So it seems that

possibly

the lower 66hz band (say 250 hz to 312 hz) would be assigned to one

of the

lower 16 current combinations, the next 66 hz, to another

combination, etc.

for 15 current combinations for that lower pair. And then the next

pair of

contacts would span the next 969Hz, splitting it into 15 channels,

etc.

All pure conjecture on my part of course. It may be nothing like

that. "

[Guest guest]

This is very close, except it's not 8 pairs with 15 locations

between each, it is 15 pairs, with 8 locations targeted between each.

Here's how the pairs are fired:

1-2 together, then

2-3 together...etc

3-4

4-5

5-6

6-7

7-8

8-9

9-10

10-11

11-12

12-13

13-14

14-15

15-16 finally fired together to complete a pulse train.

15 pairs, where electrodes 2-15 are each used twice.

Otherwise you'd miss all that space between electrodes 2 & 3, 4 & 5 etc.

There are 8 stimulation points between each pair, giving 120

available channels.

Steve

>

> We don't currently have 16 electrodes firing at the same time,

Nora.

> They fire in sequence, I believe.

>

> said it about the way that I understand it. Lots of nerve

endings

> between adjacent electrodes and, somehow, running different

currents in the

> electrodes stimulates one (or more?) of the in-between nerve

endings.

> Different combinations of current stimulate different nerve

endings. It

> would seem (guessing) that with 8 pairs of adjacent electrodes, 15

> different combinations of current for each pair would give access

to 120

> different nerve endings or groups of nerve endings. I doubt that

a single

> nerve ending can be targeted.

>

> I think it would be fascinating to read an in depth description of

the

> process but I doubt that AB will be doing that for awhile yet.

>

> Virg

>

>

> From: " lil_nora77 " <lil_nora77@...>

>

> Seriously Virg,

>

> I do not understand the excerpt--I got completely lost at what she

is

> saying. Okay let me see if I understand how it works because I

know I

> probably will not get it right. But I'll keep trying hahha.

>

> Currently AB has 16 channels as there are only 16 electrode

contacts

> in the array. So if we have 16 electrodes all firing at the same

time

> in paired sequence, with current steering (what the heck is current

> steering--what does steering mean anywayz?) I would assume that

each

> electrode has series of sections/areas being stimulated, thus

creating

> multiple virtual channels which give us more pitch perceptions? Is

> this what ms.Osberger is saying below?

>

[Guest guest]

Virg,

I think it's more like 200-8,000, and it's 15 pairs, so you get

7800 Hz spread/15 regions = 520Hz spread within each pair.

Divide that by 8 frequency bands per pair gives you 65Hz increment

between each adjacent channel. (close to your number, but achieved

differently)

Compared that to the 350-450hz increment between each channel in

today's strategies, (eg. 200, 265, 330, 395 etc. compared to 250,

600, 950, 1300 etc.) you can see there's a lot more granularity.

Much like a digital photo with more megapixels provides crisper

details visually.

Steve

>

> > How does the audiologist doing the mapping decide which of the

120

> channels

> > to use for the 16 electrodes out of interest?

>

> I can only guess, .

>

> If the electrodes are paired, as was said, and the frequency span

overall

> is 250Hz to 8Khz, it may be something like this:

>

> 7,750 Hz spread/ 8 gives 969 Hz spread per pair. If there are 120

> channels, that would amount to 15 frequency bands per pair, each

spanning

> 1/15 of its 969Hz allotment or 66Hz per band. So it seems that

possibly

> the lower 66hz band (say 250 hz to 312 hz) would be assigned to

one of the

> lower 16 current combinations, the next 66 hz, to another

combination, etc.

> for 15 current combinations for that lower pair. And then the

next pair of

> contacts would span the next 969Hz, splitting it into 15 channels,

etc.

>

> All pure conjecture on my part of course. It may be nothing like

that.

>

> Virg

>

[Guest guest]

From: " Steve Fuchs " <SteveF_MS@...>

> This is very close, except it's not 8 pairs with 15 locations

> between each, it is 15 pairs, with 8 locations targeted between each.

Oh, of course. I didn't think it through thoroughly.

I like that better.

Tnx,

Virg