for automated fitting of cochlear implants Paul J Govaerts MSc MD PhD B Vaerenberg G De Ceulaer W Kowalczyk J Diez I Bermejo The Eargroup Antwerp Belgium amp Universities ID: 289353
Download Presentation The PPT/PDF document "Artificial intelligence" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Artificial intelligencefor automated fitting of cochlear implants
Paul J Govaerts, MSc, MD, PhDB Vaerenberg , G De Ceulaer, W Kowalczyk, J Diez, I BermejoThe Eargroup (Antwerp, Belgium) &Universities of Antwerp (BE), Leiden (NL), UNED (ES)Slide2
CI FITTING
State of the art
Comfort
based
No systematic approach, no universal GCP,
huge
variability
Tailoring from the start
Eargroup approach
Outcome
based
Systematic
approach
Start with “One fits all”, postpone tailoringSlide3
Fitting for performance
Measure outcomeInterpret MAP & outcomeModify MAPSlide4
Outcome based
intensityspectralcontent
temporal
contentSlide5
Fitting for performance
Measure outcomeInterpret MAP & outcomeModify MAP
audiometry
speech audiometry
N > 60
N > 150
F
itting to
Outcome eXpert
Govaerts, et al.
Otol
Neurotol
2010; 31(6):908-18.Slide6
FOX 1.1SW opens in the background3 active maps ready to be foxed ...Slide7
FOX 1.1User interfacePassword protected
log-inUser friendly patient-selectionSW opens in the back3 active maps ready to be foxed ...Slide8
FOX 1.1Typical procedureOpen FOX – Select MAP
Perform 2 outcome measures (20 ‘)Request advice – Judge – Accept recommendationsPut new map in processorSlide9
Case1: 4 months after switch-on
CaseSlide10
switch-on
P50P75
P25Slide11
Fitting scheme
0
1
2
3
4
5 6 7 8 9 10 11 12 18 24
AutoMaps
Switch
on
Silver
2Silver 3Silver 1Ivory 1Ivory 2Ivory 3
Gold 1
Gold 2
Gold 3
5’+40’Slide12
0
1
2
3
4
5 6 7 8 9 10 11 12 18 24Fitting scheme
AutoMaps
Switch
on
Silver
2Silver 3Gold 1Silver 1Gold 2Gold 3
Ivory
1
Ivory
2
Ivory
3
Audiogram
A§E
phoneme
discrimination
Source
MAP
Gold 3#1
Map modifications
A§E
Loudness
Scaling
Speech
Audiogram
Gold 3#2, …
5’+40’
15’
30’
30’Slide13
Fitting
scheme
5’+40’
15’
30’
30’
3
hours
0
1
2
3 4 5 6 7 8 9 10 11 12 18 24Slide14
0
1 2 3 4 5 6 7 8 9 10
11
12
18
24Vaerenberg, et al. Int J Audiol 2011; 50:50-8.Preliminary results
5’+40’
15’
30’
30’
3
hours
Switch
on
: N=8,
Fox 1.1
(EG0910)
3
months
postop
(2,5
hours
)
Ongoing
trial
Europe
, IndiaSlide15
FOX
European Multicentric StudyAndreas Büchner, Thomas Lenarz, MHH, Hannover, Germany Rolf-Dieter Battmer, Romy Goetze, UKB, Berlin, GermanyIsabelle Mosinier, Stephanie Borel, Beaujon, Paris, FranceHuw Cooper, Claire Fielden, University hospital, Birmingham, UKZebunissa Vanat, Joanne Muff, Adenbrookes, Cambridge, UK
Terry Nunn, Anzel Britz,
Guy’s
and
St.Thomas
’, London, UK
Filiep Vanpoucke, Advanced Bionics
EuropeDzemal Gazibegovic, Advanced Bionics EuropePaul Govaerts, Eargroup, Antwerp, BelgiumSlide16
Preliminary results
intensityspectralcontent
temporal
content
~ 36000 outcome
points
in 275 CI
users in 15 CI centresSlide17
Preliminary results: Audiogram
Target = 30 dB (35 for 250 Hz)Tolerance = 40 dBSlide18
AudiogramSlide19
AudiogramSlide20
Preliminary results
intensityspectralcontent
temporal
content
~ 36000 outcome
points
in 275 CI
users in 15 CI centresSlide21
Spectral discriminationSlide22
Spectral
discriminationSlide23
Spectral DiscriminationSlide24
Spectral DiscriminationSlide25
Spectral discriminationSlide26
m
-zv-z
ɛ-a
z-s
y-i
ə-ɛSlide27
Preliminary results
intensityspectralcontent
temporal
content
~ 36000 outcome
points
in 275 CI
users in 15 CI centresSlide28
Loudness
scalingSlide29
Loudness ScalingSlide30
Preliminary results
intensityspectralcontent
temporal
content
~ 36000 outcome
points
in 275 CI
users in 15 CI centresSlide31
Speech AudiometrySlide32
Speech AudiometrySlide33
Speech AudiometrySlide34
Overall (41 outcome
points)Slide35
Conclusions
Measure performance
audiometry
speech audiometry
http://otoconsult.com
Feasible
in
daily clinical practice (<10’ per test) Language independent Target = normal valuesArtificial IntelligenceAssists the audiologist to navigateOptimises resultsSystematises procedureAllows for quality control