AHS
June 2010
Lake Como
The audiometric threshold:
measured in-situ,
automated, and by the
hearing aid
Gitte Keidser, Harvey Dillon, Anna
O’Brien, Dan Zhou, Lyndal Carter,
Ingrid Yeend, and Lisa Hartley
National Acoustic Laboratories
National Acoustic Laboratories, Sydney, Australia
Keidser et al.
Background

Conventional audiometry

Most modern hearing aids have on-board
generators that produce pure tones  in-situ
audiometry

Low-cost tests of hearing over the telephone or
internet in demand  automated audiometry
National Acoustic Laboratories, Sydney, Australia
Keidser et al.
Self-fitting hearing aid (concept)
Real-ear to coupler
difference
Adjust
Hearing Aid
Prescription
Formula
Automatic
Audiometer
 meet high demands in
developing countries
where services are scarce
National Acoustic Laboratories, Sydney, Australia
Keidser et al.
Questions

Is in-situ audiometry valid and reliable?
 What
is the effect of using different couplings
(open vs closed domes) on in-situ threshold
measurements?

Is automated audiometry valid and
reliable?
National Acoustic Laboratories, Sydney, Australia
In-situ audiometry


24 participants with
known sensorineural
hearing loss
Measurements (twice)
 Conventional
audiometry (insert)
 Manual in-situ
audiometry (open and
closed instant tips)
 REDD (insert, open,
closed)
Keidser et al.
National Acoustic Laboratories, Sydney, Australia
Thresholds in dB HL
(Test-retest, p = 0.69)
Keidser et al.
National Acoustic Laboratories, Sydney, Australia
REDD values
(Test-retest, p = 0.58)
p < 0.0000001
Keidser et al.
National Acoustic Laboratories, Sydney, Australia
Thresholds in dB SPL
dB SPL = dB HL + REDD
Keidser et al.
National Acoustic Laboratories, Sydney, Australia
Reliability
Keidser et al.
National Acoustic Laboratories, Sydney, Australia
Keidser et al.
Reliability
Supra-aural
Valente et al
Hawkins et al
National Acoustic Laboratories, Sydney, Australia
Keidser et al.
Reliability
Supra-aural
Valente et al
Hawkins et al
Insert
Saunders &
Morgan
Valente et al
Hawkins et al
Keidser et al.
National Acoustic Laboratories, Sydney, Australia
Automated audiometry
-10


 Manually
 Automated,
procedure
adaptive
0
10
20
30
40
50
Threshold (dB HL)
23 participants (46
ears)
Threshold
measurements at
0.25, 1, and 4 kHz
60
70
80
250
1000
Frequency (Hz)
4000
AA05
AA06
AA07
AA08
AA09
AA11
AA12
AA13
AA14
AA15
AA16
AA17
AA18
AA19
AA20
AA21
AA22
AA23
AA24
AA01
AA02
AA03
AA04
AA05
AA06
AA07
AA08
AA09
AA11
AA12
AA13
AA14
AA15
AA16
AA17
AA18
AA19
AA20
AA21
AA22
AA23
AA24
National Acoustic Laboratories, Sydney, Australia
Validity
Keidser et al.
National Acoustic Laboratories, Sydney, Australia
Keidser et al.
Reliability
Manual audiometry (5 dB
Hughson-Westlake)
Automatic audiometry (2 dB
final step size)
Automated audiometry is at least as accurate as
manual audiometry
National Acoustic Laboratories, Sydney, Australia
Keidser et al.
Conclusions

Is in-situ audiometry valid and reliable?

 corrections
for coupling required (REDD)
 ambient noise is controlled
 open tips more reliable than closed tips

Is automated audiometry valid and
reliable?

Is automated in-situ audiometry valid and
reliable? Most likely

National Acoustic Laboratories, Sydney, Australia
Keidser et al.
Acknowledgement
The in-situ audiometry study was partly
sponsored by Siemens Instruments
 The automated audiometry study was
partly sponsored by Australian Hearing
 The self-fitting hearing aid concept is now
being further explored through the Hearing
CRC

Thank you for listening!
For further information:
www.nal.gov.au
www.hearingcrc.org
Gitte.keidser@nal.gov.au