1. linical & Experimental Audiology Speech-in-noise screening tests by internet; improving test sensitivity for noise-induced hearing loss Monique Leensen Wouter Dreschler Academic Medical Centre, Amsterdam, The Netherlands m.c.leensen@amc.nl

2. linical & Experimental Audiology 2 27-2-2010 Screening  Early detection hearing impairment  Raising awareness about hearing loss  Self-assessment screeningtests: National Hearing Test by telephone (Smits et al. ’04) Online speech-in-noise tests  Easy to administer Introduction

3. linical & Experimental Audiology 3 27-2-2010 Online speech-in-noise tests  Adjustable presentation level  Spectrum of masking noise is equal to speech spectrum Independent of presentation level  Adaptive procedure  Fixed noise level  Speech Reception Treshold (SRT) 50% correct word recognition Introduction

4. linical & Experimental Audiology 4 27-2-2010 Oorcheck (LUMC ’04)  Target population: youngsters (12-24 years)  Closed set of 9 CVC words Introduction

5. linical & Experimental Audiology 5 27-2-2010 Oorcheck and NIHL  Aims at youngsters Risk for hearing loss after exposure to loud music Raising awareness about hearing loss  However, NIHL does not greatly affect SRT in stationary noise (Smoorenburg 1992) Closed set of speech stimuli Hearing threshold levels at low and middle frequencies are (near) normal Introduction

6. linical & Experimental Audiology 6 27-2-2010 Objectives of this study 1.Investigate the potential of Oorcheck to discover NIHL in an early stage 2.Examining ways to enhance the test sensitivity for mild NIHL using different forms of noise modifications  Make existing speech-in-noise screening tests (like Oorcheck) applicable for NIHL Objectives

7. linical & Experimental Audiology 7 27-2-2010 Objectives of this study 1.Investigate the potential of Oorcheck to discover NIHL in an early stage 2.Examining ways to enhance the test sensitivity for mild NIHL using different forms of noise modifications  Make existing speech-in-noise screening tests (like Oorcheck) applicable for NIHL Objectives - 1

8. linical & Experimental Audiology 8 27-2-2010 Subjects n=100 (2 excluded) NH: 49 (age 27 yrs ± 8.5) with HTLs ≤ 15 dB HL HI: 49 (age 56.3 yrs ± 9.4) - 25 with narrow dip (ND) - 24 with broad dip (BD) HTL 2kHz affected Methods - 1

9. linical & Experimental Audiology 9 27-2-2010 Test sensitivity for NIHL Sensitivity 51%  49% HI is classified as NH Specificity 90%  10% NH is classified as HI Results - 1

10. linical & Experimental Audiology 10 27-2-2010 Objectives of this study 1.Investigate the potential of Oorcheck to discover NIHL in an early stage 2.Examining ways to enhance the test sensitivity for mild NIHL using different forms of noise modifications  Make existing speech-in-noise screening tests (like Oorcheck) applicable for NIHL Objectives - 2

11. linical & Experimental Audiology 11 27-2-2010 Noise modifications  Spectral filtering Low pass filtered noise  Less masking of high frequencies  Temporal modulation NH benefit more from interrupted noise than HI (e.g. De Laat & Plomp 1983, Festen & Plomp 1990, Versfeld & Dreschler ‘02)  Both types of modifications combined  Noise parameters were based on model predictions (ESII) and pilot measurements Methods - 2

12. linical & Experimental Audiology 12 27-2-2010 Experimental set of noises test noisefilteringmodulationnoise floor 1Oorcheck--- 216 Hz- -15 dB 3LPLP 1.4 kHz--15 dB 4LP & 16 HzLP 1.4 kHz16 Hz-15 dB 5HPHP 1.4 kHz--15 dB 6HP & 16 HzHP 1.4 kHz16 Hz-15 dB 7OC -15---15 dB LPHP Methods - 2

13. linical & Experimental Audiology 13 27-2-2010 Results – group differences Results - 2

14. linical & Experimental Audiology 14 27-2-2010 Results – group differences Results - 2

15. linical & Experimental Audiology 15 27-2-2010 Results – group differences Results - 2

16. linical & Experimental Audiology 16 27-2-2010 Results – group differences Results - 2

17. linical & Experimental Audiology 17 27-2-2010 Results – group differences Results - 2

18. linical & Experimental Audiology 18 27-2-2010 Results - frequency dependence Correlations between individual SRTs and HTLS  Highest correlation for LP noise and 2-6 kHz Results - 2

19. linical & Experimental Audiology 19 27-2-2010 Validation of OC LP Increased sensitivity & specificity:  sensitivity 95%  specificity 98% Results - 2

20. linical & Experimental Audiology 20 27-2-2010 Discussion  To be investigated further: Extrapolation of lab results  ambient noise  system settings  transducers Level dependency Discussion

21. linical & Experimental Audiology 21 27-2-2010 Conclusions 1. The original Oorcheck showed only small differences between NH and HI subjects with mild NIHL (narrow dips) 2. Hearing-impaired subjects deviated more clearly from normal performance when using low-pass filtered masking noise  This suggests that this noise condition is more sensitive to discover NIHL in an early stage Conclusions

22. linical & Experimental Audiology 22 27-2-2010 Generalization We expect that the approach used will be valid for other types of speech-in-noise tests as well If this can be proven, manipulation of the masking noise will increase the sensitivity and specificity with respect to NIHL for different types of speech-in-noise screening tests Conclusions

23. linical & Experimental Audiology 23 27-2-2010 23 Thank you! Questions?

24. linical & Experimental Audiology 24 27-2-2010

25. linical & Experimental Audiology 25 27-2-2010