Download Combining Cochlear Implants and Hearing Instruments

Combining Cochlear Implants and
Hearing Instruments
Norbert Dillier
Laboratory of Experimental Audiology
ENT Department University Hospital Zurich,
Switzerland
3rd International Pediatric Conference “A Sound Foundation through Early Amplification”,
Chicago, November 4 – 6, 2004
Clinical observations
• “…an increasing number of people with
unilateral cochlear implants have residual
hearing in the nonimplanted ear that is
usable with acoustic amplification.
• Nevertheless, many adults who receive a
unilateral cochlear implant do not continue
to wear a hearing aid in the nonimplanted
ear.”
Ching et al., Ear and Hearing 25/1 (2004)
Cochlear implants & hearing instruments
• Differences and similarities
• Limitations of cochlear implants
• Additional benefits with hearing
instruments (bimodal stimulation)
• Bilateral cochlear implants
• Electroacoustic stimulation (EAS)
Cochlear Implants vs. Hearing Instruments
differences
Indication
Total deafness
Signal processing
Coding strategies
Fitting procedures
Evaluation
Objective measures
(neural vs. acoustic
responses)
Psychoacoustic and
electrophysiological
procedures
Cochlear Implants vs. Hearing Instruments
similarities
Indication
Profound deafness
Signal processing
Multichannel compression
Loudness equalization
Fitting procedures
Software based
Fitting rules
Evaluation
Free field measurements
Speech recognition tests in
quiet and noise
Cochlear Implant Principles
• Selective stimulation of small groups of
nerve fibers (with minimal channel
interaction)
• Adequate and suitable signal processing
strategies
• Appropriate mapping of signal parameters
to patient-specific psycho-electrical
stimulation conditions (loudness, pitch,
timbre)
Cochlear Implant Research
•
•
•
•
•
Electrodes, implanted electronics
Speech coding strategies
Fitting procedures
Preprocessing, noise reduction
Bilateral/bimodal stimulation
Components of a Cochlear Implant System
Transmitter coil
Receiver/Stimulator
Microphone
Speech
processor
Electrodes
Implant
Patient variables which may
influence choice of coding strategy
• Etiology
• Age, duration of
deafness
• Habituation, training
• Distribution and
condition of auditory
nerve fibers
Intracochlear recordings of electrically evoked compound
action potentials (ECAP)
Stimulus
Neural response
Auditory nerve
Implanted electrodes
Neural response telemetry recordings
• Assess peripheral function of auditory pathway
(not central mechanisms!)
• Can be obtained intraoperatively
• Can be repeated any time even with
uncooperative, sleeping or (hyper)active subjects
• Can be performed as a simple, semi-automatic
procedure for routine clinical use or using more
sophisticated, research-oriented paradigms
NRT Data and Speech Processor Maps
•
Several studies have shown that the
Neural Response threshold often
lies within the T- and C-levels of
the patient’s Map
•
The profile across the electrode
array of the Neural Response
thresholds are also often
parallel to the profile across
the array of the T- and C-levels
of the Maps
•
This may help us to
– Locate the operating range
of the Map quickly and
– Set the Map profile across
the array with only a small
amount of data
220
SB
200
NRT (6 Wks after
first TuneUp)
C-Level
180
160
T-Level
140
. 22 21 20 19 18 17 16 15 14 13 12 11 10 9
8
7
6
5
4
3
2
1
.
Benefits of performing intraoperative measurements
• NRT data is available prior to Speech Processor
fitting
• NRT measurements can
be made in the OR while
the surgeon is finishing
the operation
• Data for estimating the
Neural Response threshold
for all 22 intracochlear
electrodes can be collected
within less than 10 minutes
in the Operating room using
automated procedures
NRT threshold estimation
175
Speech processor fitting
• NRT measurements can provide valuable information
for speech processor map generation
• Behavioral thresholds vary and change over time.
• Stimulation at higher rates may decrease NRT
amplitudes and elevate NRT thresholds
• This needs to be considered
when introducing fitting
procedures based on NRT
into clinical practice
Sound pattern
is converted into a
stimulation pattern
Speech coding strategies
• SPEAK
Spectral Peak
– Low pulse rate (250 pps)
with 6 to 10 spektral maxima
• CIS
Continuous Interleaved Sampling
– High pulse rate (2000 pps) on
fixed stimulation channels
• ACE
Advanced Combination Encoder
– High pulse rate with many
spectral maxima
SPEAK (spectral peak)
„a“
6-9 maxima
20 frequency bands
4 ms = 250 Hz
Continuous Interleaved Sampling CIS
„a“
4-12 maxima / frequency bands
720-2400 Hz
1/f
Advanced Combination Encoder ACE
„a“
6-16 maxima
22 frequency bands
500-2400 Hz
1/f
Audio simulations
SPEAK
Electrodes
100%
scheuen: speak
Electrodes
0
ACE
250
scheuen: ace900
Electrodes
Time (ms)
250
scheuen: cis8-900
Time (ms)
500
100%
0%
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
0
500
100%
0%
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
0
CIS
0%
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
250
Time (ms)
500
Simulation of tonotopy
Place frequency map
Characteristic frequency (Hz)
Distance from round window (mm)
0
Electrodes
10000
1
-5
0.75 mm
-10
17 mm
1000
Speech
frequency
region
-15
22
-20
-25
100
-30
Speech sounds: formant frequencies
F1
F2
Frequency map
7938
188
F2: 15 electrodes
Frequency resolution for vowel perception
4 – 5 electrodes
F1: 8 – 10 electrodes
Piano keyboard – frequency resolution
2-3 electrodes
261.6 Hz
523.2 Hz
F2: 15 electrodes
Frequency resolution for
speech and music
4 – 5 electrodes
F1: 8 – 10 electrodes
Envelope and temporal fine structure
Z.M. Smith, B. Delgutte, A. J. Oxenham (2001)
Envelope - speech
1 ch
8 ch
32 ch
Fine structure - speech
1 ch
8 ch
32 ch
Original - speech
Envelope - music
1K
8K
32 K
Fine structure - music
1K
8K
32 K
Original - music
Conclusions
• Current cochlear implants provide good spectral
and temporal resolution for speech recognition but
insufficient resolution for music
• Increased spectral and/or temporal resolution is
needed for better music perception
• Even a small amount of additional fine temporal
structure such as that provided by a single
frequency band channel may improve music
recognition and sound quality
• Acoustic stimulation of residual hearing is one
possible way of achieving this goal
Music tests with CI and HI: pilot study subjects
M.N.G.
K.S.
C.H.
I.D.
T.B.
Pitch discrimination with CI & HI
Pitch Discrimination
100
CI & HI
CI
HI
90
80
70
[%]
60
50
40
30
20
10
0
M.G.
K.S.
C.H.
Subject
T.B.
I.D.
Melody recognition with CI & HI
Melody Recognition
100
CI & HI
CI
HI
90
80
70
[%]
60
50
40
30
20
10
0
M.G.
K.S.
Subject
C.H.
Sound quality judgements: subject TB
T.B. 10.6.04
30
2 - CI & OWN HI
3 - CI
25
1 - CI & NEW HI
Relative Score
20
15
10
5
0
Sounds like
music
Pleasant
Natural
Clear
Easy to follow
Hearing with two ears:
bilateral Cochlear Implantation,
bimodal or electroacoustic hearing
Additional mechanisms
• Interaural Level Differences (ILD)
• Interaural Time Differences (ITD)
Interaural Level differences (ILD)
Interaural Time Differences (ITD)
Bilateral cochlear implant
multicenter study
• Collaborative study with 12 German language centres
in Germany, Austria and Switzerland (Laszig et al.)
• 30 adult bilateral subjects, implanted simultaneously or
sequentially
• Analysis of unilateral vs bilateral performance on
speech perception and localisation measures
• Localization measures
• Investigation of Advanced Signal Processing
(Beamforming and bilateral noise supression
algorithms).
Bilateral Head Shadow Benefit
• Benefit of bilateral stimulation by switching to the ear
with the better signal to noise ration for spatially
separated sound sources
NL or-45°
SL
S
SR or
45°NR
N
Difference between result in
SR&NL and NR&SL condition
for each ear:
HSR = SR&NL – NR&SL
HSL = SL&NR – NL&SR
Binaural Unmasking (Squelch)
• The benefit from using two ears compared to the
ear1 with the better signal to noise ratio alone,
with spatially separated signal and noise.
SL
NR
-45°
S
45°
N
Example:
Difference between left ear alone and
binaural condition.
1in
case of asymetric hearing, the ear with the better SNR shall be the better ear.
Head Shadow Benefit und Squelch: OLSA
Oldenburg Sentencs (OLSA)
-25
SNRcrit (dB)
-20
-15
-10
-5
0
5
10
HSR
HSL
Squelch NR
Squelch NL
Dillier et al. (2003) DGA-Jahrestagung, Würzburg
Localization experiments: test setup
11
12
1
10
2
1.5 m
9
8
3
4
7
5
6
Monaural and binaural localization
(subject TN)
Binaural TN
right TN
8
8
7
7
7
6
6
6
5
4
3
Localization error
8
Localization error
Localization error
left TN
5
4
3
5
4
3
2
2
2
1
1
1
0
0
1
2
3
4
5
6
7
8
Loudspeaker
9
10 11 12
0
1
2
3
4
5
6
7
8
Loudspeaker
9
10 11 12
1
2
3
4
5
6
7
8
Loudspeaker
9
10 11 12
Monaural and binaural localization
(subject RW)
Binaural RW
Right RW
8
8
7
7
7
6
6
6
5
4
3
Localization error
8
Localization error
Localization error
Left RW
5
4
3
5
4
3
2
2
2
1
1
1
0
0
1
2
3
4
5
6
7
8
Loudspeaker
9
10 11 12
0
1
2
3
4
5
6
7
8
Loudspeaker
9
10 11 12
1
2
3
4
5
6
7
8
Loudspeaker
9
10 11 12
Summary bilateral study
• Bilateral benefits demonstrated for
headshadow effect (~10 dB SNR)
• Bilateral advantage shown in all conditions
(~ 10 dB for spatially separated and ~5 dB
for S0N0)
• No binaural squelch effect found
• Significant benefit for localization abilities
in bilateral condition
• Bilateral implantation has captured the best
hearing ear.
Bimodal/electroacoustic Stimulation
• CI on one side, hearing instrument on
contralateral ear
– Ear with CI may be completely deaf
– Combination of auditory sensations occurs in
the central auditory system
• CI and hearing instrument on the same ear
– CI surgery should not impair the preoperative
residual hearing (“safe surgery”)
– Combination of auditory sensations may take
place either peripherally or centrally
Sentence recognition in noise
with CI and CI+HI
speech
Satztest
-60°
HI
60°
noise
CI
Ching, T. Y. C. , Incerti, P. , Hill, M.: Binaural benefits for adults who use hearing AIDS and cochlear implants in opposite ears.
Ear and hearing 2004;25;1:9-21
Sentence recognition in noise
with CI and CI+HI
speech
-60° &
HI
noise
60°
CI
Ching, T. Y. C. , Incerti, P. , Hill, M.: Binaural benefits for adults who use hearing AIDS and cochlear implants in opposite ears.
Ear and hearing 2004;25;1:9-21
Functional performance in everyday
life – adult studies
• Sound heard in both ears rather than through
implant alone, more “natural”, own voice quality
improved
• Hearing aid picks up additional information, gives
“clarified” hearing, localization ability
• Identify speakers in shops and restaurants
• More confident in everyday life
• Can recognize songs, more speech-like sound
quality
Armstrong et al., 1997, Blamey et al., 1997, Tyler et al, 2002, Syms III
& Wickesberg, 2003, Ching et al, 2004, Hamzavi et al., 2004
Functional performance in everyday
life – children studies
• Speech perceived in the middle of the head rather
than in one ear
• Additional security, localization ability
• Initiate more conversation, require less repetition,
more confident
• Speak more clearly
• Imitate voice and intonation better
• More spontaneous and responsive in conversations
• Can recognize songs, distinguish environmental
sounds, voices
Simons-McCandless &Shelton, 2000; Ching et al,2000-04
Electro-Acoustic Stimulation
in the same ear
Electrical stimulation
of high frequency
> via Cochlear Implant
Acoustic stimulation
of low frequency
> via Ipsi-lateral Hearing Aid
in the same ear
Possible synergy with
electroacoustic stimulation
HG
HGopt
CI
CI+HGipsi
CU+HGopt
100
90
80
70
60
50
40
30
20
10
0
pre/postoperative speech discrimination
J.- Kiefer, HNO-Univ.Klinik Frankfurt, 2003
MAC - results adult CI users
(CI Center University Hospital Zurich)
Reference data of adult subjects
( N=44)
100
90
80
70
60
50
40
30
20
10
M ax
M in
M edian
0
FM
NUM
VM2
VO8
CM2
C12
FRE
Combined electro-acoustic stimulation
• Performance of subjects with cochlear implants
nowadays is often higher than performance of
subjects with residual hearing fitted with hearing
instruments
• The combination of cochlear implant and hearing
instrument in opposite ears, or bimodal stimulation,
can improve speech recognition performance
• Good low-frequency hearing is not a contraindication for a cochlear implant
– But, it would be better if we could keep that lowfrequency hearing…
Multi-center EAS Study
• 13 cochlear implant clinics in Europe
– 2 Spain, 1 Italy, 3 France, 2 Belgium, 1 Swiss,
2 Germany, 1 Austria, 1 UK.
• Patients implanted with Nucleus Contour
Advance electrode array
• Patients retaining sufficient residual hearing
use ipsi-lateral hearing aid for El-Ac
stimulation
• GCP-style clinical trial
Subjects
• Adult candidate for cochlear implantation
– ≤ 30% monosyll. or ≤ 50% disyll. words
• Post-linguistic onset of severe-to-profound
hearing loss
• A minimum of 10% open set word
recognition in the ear to be implanted
• At least 3 months experience with Phonak
Aero digital hearing aids if not a consistent
hearing aid user
Sample audiogram for candidacy in
electroacoustic study
Characteristic frequencies
Characteristic frequencies – overlapping
Characteristic frequencies - gap
Pre/postoperative audiograms
10
10
10
0
0
0
-10
-10
350°
-10
-20
380°
-20
10
0
410°
-20
-20
-30
-30
-30
-30
-40
-40
-40
-40
-50
-50
-50
-50
-60
-60
-60
-60
-70
-70
-70
-70
-80
-80
-80
-80
-90
-90
-90
-90
-100
-100
-100
-100
-110
-110
-110
-110
-120
125
-120
8K125
-120
8K 125
250
500
1K
2K
4K
10
10
0
0
420°
-10
-20
250
500
1K
2K
4K
440°
-20
-30
-40
-50
-50
-50
-60
-60
-60
-70
-70
-70
-80
-80
-80
-90
-90
-90
-100
-100
-100
-110
-110
-110
-120
125
-120
8K125
-120
8K 125
4K
4K
-120
8K 125
250
500
1K
2K
4K
-20
-40
2K
2K
-10
-40
1K
1K
0
-10
-30
500
500
10
-30
250
250
420°
-10
250
500
1K
2K
4K
Preop
Postop
250
500
1K
2K
4K
8K
Fraysse B. et al. (2004), Geneva, Indianapolis
Average hearing thresholds
EUSTUD 100 (N=5)
-10
Pure-tone Threshold dBHL
0
Pre-Op Maximum
10
20
30
Mean Pre-Op
Mean Post-Op
40
50
60
Mean LF (125-1K) shift = 25dB
70
80
90
100
110
120
130
140
125
250
500
750
1000
1500
2000
3000
4000
6000
Frequency Hz
Fraysse B. et al. (2004), Geneva, Indianapolis
RESULTS ON HEARING
CONSERVATION
9 patients with clinical criteria of inclusion
•Soft surgical technique in 7 cases
¾ Total or partial preservation
: 7
•Conventional surgical technique in 2 cases
¾ Total loss of residual hearing : 2
Fraysse B. et al. (2004), Geneva, Indianapolis
EARLY RESULTS
Hearing Aids – Pre-op
Implanted ear
Bilateral
Implanted ear – 3 months
CI alone
Electro-acoustic stimulation :
• Speech scores consistently
better at 1, 2 and 3 months.
• Better aided-thresholds
• Preferred by patient
• Improved sound quality
Percent correct score
100
90
80
70
60
50
40
30
20
10
0
CI+IpsiHA
Words in
quiet 65 dB
Sentences
in noise*
* : 5 dB SNR babble
Fraysse B. et al. (2004), Geneva, Indianapolis
Preliminary conclusions
• It is possible to preserve residual hearing with the
introduction of a “long” (standard) electrode using
appropriate surgical techniques
• 400° degrees appears to be the maximum insertion
depth to preserve thresholds down to 250-500 Hz.
• Subjects obtained additional benefit from the use
of an ipsi-lateral hearing aid in quiet and in noise
• Sound quality subjectively improved with El-Ac
despite residual hearing in other ear
Fraysse B. et al. (2004), Geneva, Indianapolis
Preliminary conclusions (cont)
• Subjects using El-Ac preferred and may do better
with a frequency-to-electrode allocation which is
closer to the normal frequency-position function,
and where low-frequency information is provided
only by the hearing aid.
• Further data is required to confirm that a nonstandard frequency allocation is useful for El-Ac.
Fraysse B. et al. (2004), Geneva, Indianapolis
Summary
• Cochlear implants and hearing instruments are
similar in many respects:
– Digital signal processing algorithms
– fitting procedures
– evaluation methods
• Combination of electric and acoustic stimulation
may have synergistic effects
• Whenever possible, hearing instruments should be
utilized to complement the use of cochlear implants
• To determinte optimal fitting strategies for combined
electroacoustic stimulation requires further
investigations
Acknowledgments
Cochlear Implant Team University Hospital Zurich
www.ci-zentrum.com
• Evelyn Leitner
• Susann Baumann
• Stephan Schmid
• Meike Brockmann
• Thomas Spillmann
• Michael Büchler
• Bernd Strauchmann
• Franziska Conod
• Olegs Timms
• Erika Gruner
• Dorothe Veraguth
• Alex Huber
• Simone Volpert
• Herbert Jakits
• Benno Weber
• Christiane Kühn
• Nicole Wild
• Wai Kong Lai
Links to web pages with CI simulations
•
•
•
•
www.ci-zentrum.com
www.rle.mit.edu/apc/
www.rti.org/page.cfm?nav=91
www.hei.org/research/depts/aip/audiodemo
s.htm
• www.utdallas.edu/~loizou/cimplants/
• www.kuleuven.ac.be/exporl/Lab/Members/
Laneau/SoftWare.htm