Accurate and practical methods for fitting non-linear hearing aids to infants and children: RECD, REAG, NAL-NL1
Teresa Ching, Louise Britton, Harvey Dillon, Mandy Hill
National Acoustic Laboratories, Australia
Overview
1. Non-linear amplification for children
2. Principles of the RECD / REAG / CG method 3. Methods for measuring RECD 4. Methods for prescribing and adjusting aids
Non-linear amplification for children/ infants?
Would children benefit from compression?
• •
Compression limiting
- reduces discomfort and distortion
Wide dynamic range compression
- reduces need for volume control adjustment - increases audibility of soft sounds - reduces risk of noise induced hearing loss
•
Different gain-frequency responses for different input levels
- maximises speech intelligibility
•
Reduces need for volume control
Linear amplification is optimised for typical input level for adults and children (Byrne &
Dillon, 1986; Byrne et al, 1990) •
Children/infants - ‘typical input levels’ are highly variable (Stelmachowicz et al, 1993) Why should children not have comfortable loudness more often?
•
Increases audibility of soft sounds
•
Children require better signal levels than adults for optimal performance
• 25 dB for infants (Nozza et al, 1991) • 17 dB for 5-year-olds (Byrne, 1983)
•
Children have less contextual, linguistic knowledge for understanding speech
Enhances auditory awareness of sounds (Jensen &
Henningsen, 2000)
•
Maximising speech intelligibility
•
Use NAL-NL1 - aimed at speech intelligibility, with loudness no more than normal
NAL-NL1 applies similar gain-frequency response to NAL-RP for typical input level (lots of support) NAL-NL1 applies more gain for low input level and less gain at high input level as required to maximise predicted speech intelligibility
•
•
•
NAL-NL1 validated for adults to be better than loudness normalisation procedures for speech perception (Keidser & Grant, 2000)
Reduces risk of noise-induced hearing loss
•
Greatest risk comes from greatest input levels WDRC has lower gain than linear for high levels
Output dB SPL
•
linear
WDRC
65
Input dB SPL
Could compression be bad for children?
•
Possibly, because compression reduces intensity difference cues in speech But, so does compression limiting and peak clipping. Inaudible speech is even worse!
•
•
Real ear to coupler difference
Average canal SPL minus 2cc SPL
20 15 10 5 0
125 250 500 1000 2000 4000 8000 Frequency (Hz)
RECD relative to adult data
RECD re adult values (dB)
20
1
3
10
6 12 24 48
0 100
1000 Frequency (Hz)
10000
Insertion gain and Real-ear aided gain
REUG = U - F
REIG = A - U
REAG = A - F
= REAG - REUG
U
A
F
F
Unaided
Aided
Transforming IG to REAG
IG = REAG - REUG REAG = IG + REUG
2.7 kHz
6 kHz
Does the different ear resonance suggest the need for different frequency response?
No direct evidence on use of superior high frequency sensitivity; Early speech discrimination abilities of normal hearing infants draw on low frequency, intensity, duration cues; Modifications of speech to infants include extended fundamental frequency range, increased duration and intensity on content words, and slower rate of delivery; horizontal localisation.
FOR MORE INFO...
Kuhl (1987; 1992)
Transforming IG to REAG
IG = REAG - REUG REAG = IG + REUG
Adult average
2.7 kHz
6 kHz
Transforming REAG to CG
REAG = CG + RECD + MLE CG = REAG - RECD - MLE
Measuring RECD
ER3A HA1
Real Ear Analyser ER3A insert phones HA2 25
Probe microphone
Real Ear Analyser
Measuring RECD
ER3A insert phones HA2 25
Probe microphone
HA2 25
Summary
Measure
thresholds (preferably in dB SPL in
canal)
Use
non-linear procedure
Ignore
the variation in canal resonance Prescribe in terms of Real Ear Aided Gain (REAG = ear canal SPL minus field SPL) Measure Real Ear to Coupler Difference (RECD = ear canal SPL minus coupler SPL)
Calculate
coupler gain prescription for the
individual child
Adjust
aid in coupler
Adult
Hearing threshold level (dB HL)
Child
Hearing threshold level (dB HL or dB SPL) Measure individual RECD Apply prescription to derive coupler gain targets Adjust hearing aid via coupler/programmer to achieve targets
Apply prescription to derive real ear targets
Adjust hearing aid via coupler/programmer to achieve targets
Verify with real ear measurement
Questions?
Measuring RECD: results
RECD = Real ear gain – Coupler gain
30 25 20
Gain (dB)
15 10 5 0 -5 -10 Frequency (Hz) 250 500 1000 2000 4000
RECD = Real ear gain – Coupler gain
30 25 20
Gain (dB)
15 10 5 0 -5 -10 Frequency (Hz) 250 500 1000 2000 4000
RECD = Real ear gain – Coupler gain
30 25 20
Gain (dB)
15 10 5 0 -5 -10 Frequency (Hz) 250 500 1000 2000 4000
RECD
real-ear SPL measured using own earmould, coupler SPL measured using a 2 cc coupler
Real-ear-to-coupler differences
25 43 ears 20 15 10 5 0 -5 0.25 0.5 1 2 4
RECD (dB SPL)
Frequency (kHz)
RECD – repeatability
Real-ear-to-coupler differences 20
n = 19
First m eas. Repeated m eas.
15
RECD (dB)
10 5 0 -5
0.25
0.5
1 Frequency (kHz)
2
4
Measuring RECD: practical issues
Left and right ears
Blocked probe tube
Leakage around earmould
Probe positioned too far away from the eardrum
Measured with active child
Leakage of probe bore into main sound bore
Fitting hearing aids using RECD
Fitting hearing aids using RECD
REAG
= RECD + CG Derive individualised coupler gain targets Adjust hearing aid parameters and measure in a coupler for verification
Deriving NAL-NL1 targets
Fitting a Siemens Prisma 2SP+
Fitting a Siemens Prisma 2SP+
Fitting a Siemens Prisma 2SP+
Fitting a Siemens Prisma 2SP+
Fitting a Siemens Prisma 2SP+
Fitting a Siemens Prisma 2SP+
50
Coupler gain (dB)
40 30 20 10 0 250 500 1000 2000 4000 Frequency (Hz)
NL1-55 NL1-70 NL1-80 Ach-55 Ach-70 Ach-80
Fitting a Bernafon LS16
Deriving NAL-NL1 targets
Fitting a Bernafon Smile III (LS16)
Fitting a Bernafon Smile III (LS16)
Fitting a Bernafon Smile III (LS16)
Fitting a Bernafon LS16:Speech display
Fitting a Bernafon LS16 – worst case
50
Coupler gain (dB)
40 30 20 10 0 250 500 1000 2000 4000 Frequency (Hz)
NL1-55 NL1-70 NL1-80 Ach-55 Ach-70 Ach-80
Fitting to NL1 targets for 43 ears
Fitting to NAL-NL1
50 45 40
55 dB
70 dB
80 dB
Prescribed Achiev ed
2 cc coupler gain (dB)
35 30 25 20 15 10 5 0 0.25 0.5 1 2 4 0.25 0.5 1 2 4 0.25 0.5 1 2 4
Frequency (kHz)
Summary
NAL methods for selecting hearing aids for children and infants
Measure
hearing thresholds and RECD using insert earphone & custom earmould Derive individualised coupler gain targets using NALNL1 software Adjust hearing aid and verify coupler gain targets Display amplified speech levels in the real ear Evaluate effectiveness
– functional assessment based on parents’/teachers’ observations in real life (M Hill) – Electrophysiological assessment (S Purdy)
Fine-tune
hearing aids and/or modify management practice if required.
www.nal.gov.au Teresa.Ching@nal.gov.au Louise.Britton@hearing.com.au National Acoustic Laboratories, Australia.