DSL v5.: A Presentation for the Ontario Infant Hearing Program Associates The Desired Sensation Level (DSL) Method Early development.... 198 Goal: To develop a computer-assisted electroacoustic-based procedure for the fitting of amplification in pre-verbal children. Richard Seewald, Marlene Bagatto and Sheila Moodie National Centre for Audiology The University of Western Ontario London Ontario Canada Early Development.... Speech perception (e.g., Erber & Witt 1977) Looked to the literature for studies of.. Speech perception in children Preferred Listening levels and Loudness discomfort in children History of the DSL Method How well does this work??? Version 1 was completed in 1984 (pencil and paper) The first software system (DSL v3.) was released in 199 (DSL for fitting linear gain instruments) DSL[i/o] for Windows was released in 1996 (DSL[i/o] v4. for linear and WDRC instruments) DSL m[i/o] v 5. was released to hearing instrument manufacturers in 25? Validation Studies
Preferred Listening Levels of Children Who Use Hearing Aids: Comparison to Prescriptive Targets Scollie, Seewald, Moodie and Dekok JAAA 2 Funding provided by Phonak AG Preferred Listening Levels in Children N = 18 Mean age = 1 years Mild to Profound SN hearing loss The subjects listened to average conversational speech and adjusted their VC to the level they preferred. The subjects preferred VC setting, for an average speech input, was compared to DSL & NAL prescribed settings Preferred Listening Levels in Children Preferred Listening Levels in Children Recommended Listening Level (db) 8 6 4 2 PLL / DSL Comparison DSL 2 4 6 8 Preferred Listening Level (db) PLL / DSL Comparison On average, the DSL prescribed setting was 2 db lower than the subject s PLLs The PLLs were within 5 db of the DSL prescribed settings for 7% of subjects Preferred Listening Levels in Children Preferred Listening Levels in Children Recommended Listening Level (db) 8 6 4 2 PLL / NAL Comparison NAL-RP/NL1 2 4 6 8 Preferred Listening Level (db) PLL / NAL Comparison On average, the NAL prescribed setting was 11 db lower than the subject s PLLs The PLLs were within 5 db of the NAL prescribed settings for 9% of the subjects
Studies of Aided Speech Perception and Loudness with DSL[i/o] algorithm Studies of Aided Speech Perception and Loudness with DSL[i/o] algorithm Jenstad et al. (1999) Speech perception Jenstad et al. (2) Aided loudness measures Jenstad et al. (1999) Speech perception: Mean sentence perception scores were high (>9% correct) across a wide range of speech input levels (i.e., 48 83 db SPL) Studies of Aided Speech Perception and Loudness with DSL[i/o] algorithm Jenstad et al. (2) Aided loudness measures: Aided growth of loudness was normalized for speech and other environmental sounds through the DSL[i/o] algorithm. DSL is just about right! Different Prescriptions for Adults and Children Background Currently, there are currently differing opinions regarding the electroacoustic requirements for adults vs. children Prescriptive procedures developed for adults can be applied with young children (Ching, et al, 21) Infants and young children require different prescriptive procedures (e.g., Stelmachowicz, 1991; 2, Seewald et al. 1996).
Studies.... MacRae (1986) Snik & Hombergen (1993) Laurnagaray & Seewald (25) Relationships Between the Hearing Threshold Levels, Aided Threshold Levels and Aided Speech Discrimination of Severely and Profoundly Deaf Children John MacRae NAL Report No. 17 January 1986 MacRae (1986) MacRae (1986) Subjects: 1 children with moderate through profound hearing loss. Measured the real-ear gain they were using and compared this to that prescribed by the NAL-R algorithm MacRae (1986) MacRae (1986)
Hearing Aid Fitting of Preschool and Primary School Children Snik and Hombergen (1993) Subjects: 4 Adults; 6 Children Snik and Hombergen Scandinavian Audiology (1993) Measured the preferred insertion gain as a function of the PTA hearing loss. Snik and Hombergen (1993) Snik and Hombergen (1993) Adults Children The results showed that overall the mean use insertion gain was 7 db less for the adults relative to that used by the children in this study Procedure Adult / Child Preferred Listening Levels Scollie, Seewald, Cornelisse, Moodie, Bagatto, Laurnagaray, Beaulac & Pumford Trends in Amplification (25) Subjects (n=72) 24 children who were full-time hearing aid users 24 adults with experience wearing hearing aids 24 adults who were new hearing aid users Prescriptive targets DSL[i/o] 4.1 Participants had a variety of nonlinear hearing instruments New users provided with 15 to 2 day period of hearing instrument use
Procedure Adult/Child Preferred Listening Levels Volume control wheel set to minimum Speech (RMS 6 db SPL) via a loudspeaker in sound field Participant set VCW until talker sounded the best (repeated measure) Preferred Listening Level 7 6 5 4 3 2 1 Children New Adult Users Experienced Adult Users 1 2 3 4 5 6 7 Recommended Listening Level Adult/Child Preferred Listening Levels Adult/Child Preferred Listening Levels Repeated Measures: The average test-retest difference was 1 db for all three groups and no participant varied by more than 6 db on repeated measures. The three groups differed in their PLLs: Children: The mean PLL was 2 db below the DSL 4.1 target Experienced Adults: The mean PLL was 9 db below the DSL 4.1 target New Adult Users: The mean PLL was 11 db below the DSL 4.1 target DSL v5.: Hearing Aid Prescription for Adults with Acquired Hearing Loss DSL[i/o] 4.1 is too loud!
DSL v5.: adult targets for range of speech inputs Ear Canal Level (db SPL) 14 13 12 11 1 9 8 7 6 5 4 3 2 1 74 7 6 52 Preferred Listening Levels (PLLs) of Adults and Children Who Wear Hearing Aids Compared to DSL v5. Target Criteria Seewald, Moodie, Scollie, Moodie & Polenko (2?) 1 1 1 Frequency (Hz) Adult PLL / target comparisons study PLL / target comparisons N = 9 adults with sensory hearing loss MEASURES: 2 15 DSL 4.1 DSL 5. 1. RECD measures 2. Set hearing aid in S-REM using DSL v5 target criteria for REAR and RESR 3. Set VCW to minimum, ask client to set to PLL using running conversational speech 4. Measure S-REM at PLL setting 5. Repeat 6. Make desired electroacoustic changes per client report 7. Measure S-REM at final setting 8. Compare deviation from PLL Deviation from PLL (db) 1 5-5 Average deviation from PLL 5 4 Hz: -1 DSL 4.1 = 12.1 db -15 DSL 5. =.5 db -2 1 1 1 Frequency (Hz) Implications for DSL v5. Now that s more like it!!! DSL v5. acknowledges that adults and children not only require, but also prefer different listening levels, thus: Thus, DSL v5. is the first generic hearing instrument prescription algorithm to prescribe different sensation levels of amplified speech for children with congenital or early onset hearing loss versus adults with late onset hearing loss.
DSL v5.: The New DSL Method for Hearing Instrument Fitting The DSL Method v5. AUDIOMETRIC ASSESSMENT Compatibility with ABR Assessment Data Updated RECD Norms ELECTROACOUSTIC PRESCRIPTION HEARING AID VERIFICATION The DSL m[i/o] Algorithm Multi-channel Targets for Quiet & Noise Modifications for Conductive Losses and Binaural Fittings Targets for Children & Adults Multi-Level Targets The DSL Method v5. AUDIOMETRIC ASSESSMENT Compatibility with ABR Assessment Data Updated RECD Norms Updated RECD measurement procedure Compatibility with ABR Assessment Data nhl to ehl Diagnostic Audiologists correct ABR data by applying frequency-specific values Data reported in ehl on ISCIS form ehl data entered into DSL software for hearing aid fitting Application in DSL v5. Three options: Enter nhl value and use DSL v5. default corrections Enter nhl value and use own correction values Enter ehl values This assumes correction has already been applied
Setup Ontario Infant Hearing Program Corrections 5 1 2 4 AC 15 1 5 *Equipment, parameter, and calibration specific Updated Average RECD Values Updated RECD Predictions for Infants Real-Ear-to-Coupler Difference (RECD) Predictions as a Function of Age for Two Coupling Procedures Marlene Bagatto, Susan Scollie, Richard Seewald, Shane Moodie & Brenda Hoover 22, JAAA, vol 13(8)
Predicted RECDs Predicted RECD Values DSL v4.1 predicted values: - foam tip coupling only - based on 12 month age ranges Newly developed predictions: - foam tip and earmold coupling - to the nearest month Subjects - 392 infants & children - ages 1 month to 16 years - 141 ears used immittance tips - 251 ears used earmolds - normal otoscopic and immittance findings Results - high variability in RECD measures for children of the same age Predicted RECD values: Earmolds Updated RECD Predictions Foam / immittance tips Earmolds 3 25 i) 3 25 ii) 2 2 RECD (db) 15 1 5 RECD (db) 15 1 5-5 -5-1 2 4 6 8 1-1 5 1 15 2 25 Age (months) Age (months) Predicted RECD values: Earmolds Predicted RECD Values Limitations: all subjects had normal middle ear function high variability in RECD measures associated with children of the same age Therefore, whenever possible, predicted values should NOT replace the more precise RECD measurement.
Comparison of RECD Norms New IHP Procedure RECD (db) 3 25 2 15 1 RECD Norms: 2 month old Old Norm New Norm If have DSL v5., use of the table of RECD values is no longer necessary Choose foam tip or earmold coupling 5 1 Frequency 1 (Hz) 1 Measured RECDs are always preferable The DSL Method v5. The DSL Method v5. AUDIOMETRIC ASSESSMENT Compatibility with ABR Assessment Data Updated RECD Norms ELECTROACOUSTIC PRESCRIPTION The DSL m[i/o] Algorithm Multi-channel Targets for Quiet & Noise Modifications for Conductive Losses and Binaural Fittings Targets for Children & Adults ELECTROACOUSTIC PRESCRIPTION The DSL m[i/o] Algorithm Multi-channel Targets for Quiet & Noise Modifications for Conductive Losses and Binaural Fittings Targets for Children & Adults HEARING AID VERIFICATION Multi-Level Targets The New DSL Multi-Stage Input-Output Algorithm The DSLv5. m[i/o] algorithm: INPUT (db SPL re: Free Field) Expansion Stage Linear Stage WDRC Stage Limiting Stage Targets for OUTPUT (db SPL in the Ear Canal)
The DSL m[i/o] algorithm 14 WDRC: Compression Threshold 8 Output db RE (SPL) 12 Upper Limit of Comfort 1 8 6 ` Threshold 4 2 Input range for WDRC 2 4 6 8 1 12 14 Input db SF (SPL) Compression Threshold (db SPL) 7 Loud Speech Average Speech 6 Soft Speech 5 4 3 2 In DSL v5., a channel-specific compression threshold prescription recommends a low compression threshold for milder hearing losses. The prescribed CT is about 5 db SPL for moderate hearing losses, and is 1 gradually raised as hearing level increases, to about 7 db SPL for severe to profound losses. 2 4 6 8 1 db HL Working with Multichannel Hearing Aids SELECTION Digital Processing Multichannel Compression DSL H E A R I N G A I D V E R I F I C A T I O N Modifications for Conductive Hearing Loss Will increase the upper limit of comfort by 25% of the air-bone gap. Resulting targets are slightly more linear, and have slightly more gain. Targets are not allowed to exceed 14 db SPL (ear canal). Modifications for Conductive Hearing Loss Ear Canal Level (db SPL) 14 13 12 11 1 9 8 7 6 5 4 3 2 1 1 1 1 Frequency (Hz) Targets for Aided Speech with 15 db conductive hearing loss Targets for Aided Speech DSL 5.: Binaural correction Ear Canal Level (db SPL) 14 13 12 11 1 9 8 7 6 5 4 3 2 1 1 1 1 Frequency (Hz) Targets for Aided Speech for monaural fitting Targets for Aided Speech for binaural fitting 3 db
Binaural Correction Notes MPO Targets The literature is not conclusive on whether a gain reduction is needed for binaural fittings in children. This element of DSL v5. requires further investigation. Until conclusive results are available, it is recommended that IHP associates choose NO binaural correction even for those children fit with binaural aids. The ULC target ( ) is really intended to be matched by fully saturated hearing aid responses, therefore a slightly lower target may be more appropriate for use with the MPO test signal. For this reason, the target input/output function within DSL v5. can be used to compute a level-dependent target for either 85 db SPL (in the real ear) or 9 db SPL (in the coupler, using simulated real ear measurement). This new target will be somewhat lower than ULC for most hearing losses. MPO Targets DSL v5: Comparison to DSL 4.1 Target Criteria * * * * * * * * * * * * * * * * * *
For Infants For Young Children For Adults
Child Care Settings are Noisy Learning Environments Seewald: Signal-to-Noise Considerations Noise levels in occupied preschool classrooms in child care centers can range from 66 to 94 db(a) Noise levels in occupied infant/ toddler classrooms range from 58 to 68 db(a) and the SNR ranges from -3 to 12 db. up to 5% of young children in child care have chronic problems with fluid in their middle ear, where they experience this condition almost 5% of the time during the first 3 years of life If one believes in an auditory approach to speech and language development then the following must occur: 1. The desired signal must be delivered within the child s residual hearing capacity. 2. The speech signal must be perceived in order for the normal course and sequence of speech and language development to occur. 3. The speech signal cannot be constantly buried in a background of noise. Manlove, Frank & Vernon-Feagans (21). DSL v5: Prescriptions for Noisy Environments Prescription for Noisy Listening Environments 14 13 12 Importance-based reduction in gain Loudness comfort Speech Intelligibility Derived from the Speech Intelligibility Index (ANSI, 1997).8.7.6.5.4.3.2.1 1 1 1 Frequency (Hz) Importance Ear Canal Level (db SPL) 11 1 9 8 7 6 5 4 3 2 1 Targets for Quiet Targets for Noise 1 1 1 Frequency (Hz) Prescription for Noisy Listening Environments The Noisy Listening Environment targets require validation It is not a minimum implementation requirement for manufacturers in this release of DSL v5 Many manufacturers, including Audioscan, have not implemented these noise targets in their first release of DSL v5 Case Examples
Outcomes 1 % vowel recognition 72% high frequency consonant recognition Loudness perception within normal range Sample verification data for hearing aid fitting to DSL v5. targets. Real-world listening rated good to great on CHILD questionnaire Evaluation of DSL v5. Preliminary Data from Children Frequency (Hz) 1 children 7 to 17 yrs 1 1 1-2 3 preferred responses slightly over target in low to mid frequencies 2 Mild to severe SNHL db HL 4 6 2 preferred responses slightly under target in low frequencies 8 Binaural fittings 55, 65, 75, MPO 1 12 5 satisfied with hearing aid set to target Loudness Outcomes DSL v5. and the SII Loudness Rating 8 7 6 5 4 3 2 1 45 5 55 6 65 7 75 8 85 Input db SPL SII summarizes the proportion of audible speech SII in Verifit based on real-ear SPL thresholds and measured REAR of the aid Overall audibility present in the fitting as a percentage
DSL v5. and SII What is a good SII? SII value does not indicate the proportion of speech that will be understood by the user SII has a level distortion factor that is applied for speech levels higher than 62.4 db SPL Speech understanding depends on development, type of speech materials, listening environment, and other factors Applied in severe to profound fittings Makes SII value appear lower than one would expect Better Ear SII (%) 1 9 8 7 6 5 4 75 db SPL 3 65 db SPL 2 55 db SPL 1 2 4 6 8 1 12 Better Ear HF-PTA (db HL) 75 db HL and better SII >6% 8 db HL and above SII 2 to 6% IHP Protocol Review e An individual with congenital hearing loss can be fit with DSL child targets regardless of age. Transducer is ehl Measure RECDs or choose average Binaural is set to NO IHP Verification Review IHP Verification Protocol
DSL Website Update Protocol Review Total website redesign will begin in January 27 Goal for completion: Late Spring 27 Discussion: suggestions for website Discussion re: challenges in clinical practice in following recommended IHP amplification protocol.