Paintings & Photography

Converted and Upgraded Maps Programmed in the Newer Speech Processor for the First Generation of Multichannel Cochlear Implant

Converted and Upgraded Maps Programmed in the Newer Speech Processor for the First Generation of Multichannel Cochlear Implant
of 11
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Copyright @ 2013 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited. ARTICLE COVER SHEET LWW_MAO-FLA-BRV-EDISERVER-BASED Template version : 4.2Revised: 08/10/2012 Article : MAO22355Creator : jmaxinoDate : 6/26/2013Time : 11:29 Number of Pages (including this page) : 10  Copyright @ 2013 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited. Converted and Upgraded Maps Programmed in the Newer Speech Processor for the First Generation of Multichannel Cochlear Implant  Ana Tereza de Matos Magalha ˜es, M. Vale´ria Schmidt Goffi-Gomez,Ana Cristina Hoshino, Robinson Koji Tsuji, Ricardo Ferreira Bento,and Rubens Brito  Department of Otolaryngology, University of Sa˜o Paulo School of Medicine, Sa˜o Paulo, Brazil  Objective:  To identify the technological contributions of thenewer version of speech processor to the first generation of multichannel cochlear implant and the satisfaction of users of the new technology. Among the new features available, we fo-cused on the effect of the frequency allocation table, the T-SPLand C-SPL, and the preprocessing gain adjustments (adaptivedynamic range optimization). Study Design:  Prospective exploratory study. Setting:  Cochlear implant center at hospital. Patients:  Cochlear implant users of the Spectra processor withspeech recognition in closed set. Seventeen patients were se-lected between the ages of 15 and 82 and deployed for more than8 years. Interventions:  The technology update of the speech processor for the Nucleus 22. Main Outcome Measures:  To determine Freedom’s contribu-tion, thresholds and speech perception tests were performedwith the last map used with the Spectra and the maps createdfor Freedom. To identify the effect of the frequency allocationtable, both upgraded and converted maps were programmed.One map was programmed with 25dB T-SPL and 65dB C-SPLand the other map with adaptive dynamic range optimization. Toassess satisfaction, SADL and APHAB were used. Results:  All speech perception tests and all sound field thresh-olds were statistically better with the new speech processor;64.7% of patients preferred maintaining the same frequencytable that was suggested for the older processor. The sound fieldthreshold was statistically significant at 500, 1,000, 1,500, and2,000 Hz with 25dB T-SPL/65dB C-SPL. Regarding patient’ssatisfaction, there was a statistically significant improvement,only in the subscale of speech in noise abilities and phone use. Conclusion:  The new technology improved the performance of  patients with the first generation of multichannel cochlear im- plant.  Key Words:  Adult   V  Audiometry  V  Cochlear implant   V  Conversion  V  Speech discrimination tests  V  Speech perception  V  Speech processor. Otol Neurotol   00: 00  Y  00, 2013. The first multichannel cochlear implant surgeries inBrazil were performed in 1999 usingthe Spectra body-wornspeech processor from the Cochlear Limited (Melbourne,VIC, Australia), which was developed as the internal de-vice for the Nucleus 22 (1). Over the years, new technol-ogies havebeenlaunched, and cochlearimplant companieshave created speech processors that were compatible withthe former internal device, avoiding the need for further surgery (2  Y  5).In 2006, the newer speech processor was introducedwith new technological features, such as 2 new micro- phones and a new chip for signal processing that improvesthe ability to process incoming acoustic sound not onlyfor new users but also for patients with earlier generationsof cochlear implants. Some of the innovations include newlevels for the input (T-SPL) and output (C-SPL) of acous-ticsignalsto25and65dB,respectively,andnewstrategiesto preprocess sound (adaptive dynamic range optimization[ADRO]) (6).In 2009, the replacement parts for the older processor system were no longer being manufactured, and the pro-cessor was considered obsolete. The Public Health Systemin Brazil then authorized switching all patients with thefirst generation of multichannel cochlear implant from theolder body-worn speech processor to the newer speech processors at no cost.The literature has shown that there are advantageswhen outdated speech processors are replaced by a newer device (3  Y  5,7). However, research specific to the new Address correspondence and reprint requests to Ana Tereza de MatosMagalha˜es  AQ1  , Department of Otolaryngology, University of Sa˜o Paulo Schoolof Medicine, Rua Capote Valente, 432 Conjunto 14, Sa ˜o Paulo-SP, Brazil,CEP- 05409-001; E-mail: This manuscript has no conflict of interest. Otology & Neurotology 00: 00  Y  00    2013, Otology & Neurotology, Inc. 1 Copyeditor: Lorelie Eugenio  Copyright @ 2013 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited. technological resources that could benefit the first gener-ation of cochlear implant patients could not be found.This study aimed to identify whether the technologicalcontributions of the newer version of speech processor to the first generation of multichannel cochlear implant were beneficial to experienced users of older technolo-gies. With the technological features available in the new processor, we focused on the effects of the frequencyallocation table, the T-SPL and C-SPL and the prepro-cessing gain adjustments (ADRO), as well as the overallsatisfaction of the users of the new technology. MATERIALS AND METHODS This was a prospective exploratory study to evaluate the re-sults obtained with the updated technology available in thespeech processor for the first multichannel cochlear implant  Nucleus 22.A survey was performed on all 43 patients who were im- planted with the first generation of cochlear implant from 1999to 2001. This study included teenage and adult patients who ef-fectively used the implant (at least 8 h/d) with no previous expe-rience with the new technology and had at least some speechrecognition on a closed set with their processor. Eleven patientswere no longer implant users, 4 were children, 3 had no speech perception, and 8 patients were already users of the newer pro-cessor before the beginning of this study. Seventeen patientsmet the inclusion criteria, including 9 male and 8 female subjectsranging in age from 15 to 82 years. Most of the sample consistedof patients with more than 8 years of experience with cochlear implant use. All patients were postlingually deaf except for onewith congenital deafness (  T1 Table 1).The protocol strictly followed by all patients was divided into3 steps (  F1 Fig. 1): a. First Step: Optimizing the current map and verifying theoperation of the older processor In the first stage, the final current map used by the patient with the Spectra processor was revised and optimized withCustom Sound 2.0 software (Cochlear Limited, Melbourne,Victoria, Australia) according to our programming routine. Weexamined the minimum level of stimulation (T level) at whichthe patient could identify 100% of the presentations, as wellas the maximum comfort level (level C) at which the electricalcurrent demonstrated a comfortable level of loudness. After-wards, level C was balanced. This map was designated as theSpectra map (SM). Before the tests, the operation of the Spectra  processor was confirmed, and the wires, transmitting coil, andmicrophone were checked using a signal check and an earphone TABLE 1.  Sample demographics Patients Sex Age (yr) Cause Duration of deafness (yr)Time of cochlear implantationuse (yr) N active electrodes on the map(active channels)S1 M 50 Progressive 15 8 15 (17)S2 M 68 Progressive 10 9 16 (18)S3 F 47 Chickenpox 23 9 16 (18)S4 F 63 Otosclerosis 5 10 14 (16)S5 M 82 Unknown 2 11 20 (20)S6 M 49 Traumatic brain injury 2 10 20 (20)S7 M 69 Progressive 3 9 18 (20)S8 F 43 Meningitis 1 11 16 (18)S9 F 15 Congenital 7 8 18 (20)S10 M 29 Progressive 4 10 20 (20)S11 F 47 Meningitis 28 10 20 (20)S12 F 43 Meningitis 27 8 18 (20)S13 M 58 Traumatic brain injury 2 8 18 (20)S14 M 17 Meningitis 1 8 5 (5)S15 F 30 Unknown 4 9 18 (20)S16 M 50 Unknown 11 11 20 (20)S17 F 44 Unknown 9 8 19 (19)F indicates female; M, male. FIG.1.  Summary of the steps of this research. *Sound-field audiometry tests and speech tests in an acoustically isolated booth and visual-analogue scale tests were performed. ** Live voice tests (vowels and monosyllables) and visual-analogue scale tests were performed. 2 A. T. DE MATOS MAGALHA˜  ES ET AL. Otology & Neurotology, Vol. 00, No. 00, 2013  Copyright @ 2013 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited. monitor. If a fault was identified, the component was replaced before testing. b. Second step: Programming the newer processor To identify the effect of the frequency allocation tables(FATs), the programming of the new speech processor was performed using the following 2 techniques: •  Technique 1: Converting the map that was programmedfor the Spectra, which involves maintenance of the fre-quency allocation table. The levels of stimulation werereviewed, optimized and balanced, and all other parame-ters from the srcinal map were maintained (channel andtotal stimulation rate, mode stimulation, T-SPL of 30 dBand C-SPL of 70 dB, with no pre-processing of the signal).This map was designated as the converted map (CM). •  Technique 2: Updating the map that was programmedfor the Spectra, which involves changing the frequencyallocation table to the table suggested for the Freedom(188  Y  7,938 Hz). The levels of stimulation were reviewed,optimized, and balanced, and all other parameters of thesrcinal map were maintained (channel and total stimula-tion rate, stimulation mode, T-SPL of 30 dB and C-SPL of 70 dB, with no preprocessing of the signal). This map wasdesignated as the updated map (UM).After the programming of the 2 new maps, the live voice tests performed at 60 dB SPL became more appropriate to deter-mine the preference of frequency allocation tables, as the pro-cessor was connected to the computer at this step. The perceptionof vowels and monosyllabic words were only subjected to audi-tory evaluation. The order of presentation of the converted (CM)andupdated(UM)mapswasrandomizedat the speech perception tests, each patient defined his/her  preference using a visual analogue scale (VAS). Each patient was asked to mark an X from 0 to 100 on a 100-mm line, where0 corresponds to the worst clarity and naturalness of sound and100 corresponds to the best clarity and naturalness of sound (8).The patients were questioned regarding their preferences between the 2 maps: converted or upgraded. If the patient couldnot define this difference using the VAS, the map that per-formed better in speech perception tests was chosen accor-ding to the order of difficulty, with monosyllables first andthen vowels. c. Third step: Optimization of the CM and UM maps Either the UM or CM, according to the patient’s choice or per-formance,werewritteninprogramP1ontheFreedomprocessor.Two new maps were programmed to identify the effect of T-SPL and C-SPL on pre-processing ADRO, which were writtenin programs P2 and P3, respectively.The new device programming protocols were as follows: •  Map CM-SPL or UM-SP: maintained all parameters of the srcinal map (channel and total stimulation rate), except for T-SPL of 25 dB and C-SPL of 65 dB. These maps werewritten in program P2. •  Map CM-AD or UM-AD: maintained all parameters of the srcinal map (channel and total stimulation rate, andT-SPL of 30 dB and C-SPL of 70 dB) while implementingADRO. These maps were written in program P3. Evaluation of Audiometric Thresholds and SpeechPerception Tests in the First and Third Stages Audiometry tests were performed in the sound field boothwith the older speech processor and the newer speech processor.For the statistical analysis, we considered the average audio-metric thresholds of 500, 1,000, 2,000, and 4,000 Hz (9) andthe frequency threshold separately.In the same session, after programming all the maps, speech perception tests were performed in the sound field and in anacoustically isolated booth with CD-recorded material presentedat 60 dB SPL (10): •  Monosyllables, •  sentences in an open set in silence (11), •  sentences in an open set in noise with SNR = 0 dB** this test was only applied when sentences in the open set insilence were over 50%.The list of sentences was determined using the test proposed by Costa et al. (11). This material is composed of 7 lists of 10 phonetically balanced sentences that were recorded by a profes-sional male announcer on a compact disc. The SNR of 0 dB wasused for sentences in noise because we wanted to avoid theceiling effect, and none of the patients who had closed set speech perception in silence performed worse than 10%.The order in which the SM and the new maps for the Freedom processor (P1, P2, and P3) were tested was randomized at the speech tests with each program using the newer  processor, another VAS evaluation was requested from the pa-tient for each map.To evaluate the performance of the newer processor, the re-sults achieved using the program (P1 or P2 or P3) with the best response were compared with the results of the SM. The choiceof the program with the best response was based on the order of test difficulty (monosyllables, sentences in noise, and sentencesin silence).To identify the effect of T-SPL and the C-SPL, the soundfield thresholds were compared for statistical significance. Thespeech perceptions in quiet (monosyllables) and noise (open set in noise) were obtained using maps written in P1 and P2.To identify the effect of ADRO, the performance of speech perception in silence (monosyllables) and noise (open set innoise) was compared with the speech processor programs P1and P3.To assess satisfaction with their hearing devices, 2 question-naires were used as follows: the Satisfaction with Amplificationin Daily Life (SADL) (12) and the Abbreviated Profile of Hearing Aid Benefit (APHAB) (13). SADL was applied beforeactivation and after both a month and a year of using the newer  processor, whereas APHAB was measured after 1 month andafter 1 year of activation. We chose to compare the SADL scale beforethe upgrade/conversion ofthe olderprocessortothe SADLscale 1 month and 1 year later with the newer processor because TABLE 2.  Results of speech perception with both speech processors Spectra map Best freedom map (P1 or P2)Median (P25%, P75%) Mean (SD) n Median (P25%, P75%) Mean (SD) n  z p  (Wilcoxon)Monosyllables 16 (4, 32) 20.9 (18.2) 17 36 (20, 56) 34.8 (19.9) 17  j 3.258 0.0011Sentences in noise 0 (0, 60) 26.5 (32.8) 17 40 (10, 80) 46.5 (36.7) 17  j 3.073 0.0021Sentences in quiet 20 (0, 30) 20.0 (18.7) 9 60 (60, 70) 56.7 (26.0) 9  j 2.378 0.0174 3CONVERTED AND UPGRADED MAPS FOR CI USERS  Otology & Neurotology, Vol. 00, No. 00, 2013  Copyright @ 2013 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited. we wanted to verify the level satisfaction with this new tech-nology and to assess whether satisfaction changes over time.The APHAB was used to compare the 2 programs betweenthe new device and older device. In this study, the question-naire was applied for both devices; subtracting one score fromthe other generates a measure of the benefit provided by thenew processor.Patients reported the amount of trouble they had with commu-nication or noises in various everyday situations in a self-assessedinventory that comprised 24 items scored in 4 subscales. The sub-scalesincludedthefollowing:EaseofCommunication(EC),whichmeasures the strain of communicating under relatively favorableconditions; Reverberation (RV), which assesses communicationin reverberant rooms; Background Noise (BN), which is evaluatedin settings with high background noise levels; and Aversiveness(AV),whichtalliestheunpleasantnessofenvironmentalsounds.For analysis, the mean of each subscale was subsequently subtractedto compare the newer and older conditions of the individualsubscales. A difference of approximately 22 points between the2 devices for EC or RV or BN, and 31 points for AV, was nec-essary to be reasonably certain that the change in scores repre-sented a real difference. The global score represented the 3communication subscales together. If EC, RV, and BN are supe-rior by at least 10 points, you can be fairly certain that the better scoring is truly superior.The SADL is an inventory cluster of 15 items set into 4subscales: positive effects (evaluation of acoustic and psycho-logical benefits), negative features (background noise, feed- back, and telephone use), cost and service, and personal image(cosmetics and hearing aid stigma). A value for each of the 15items was obtained by assigning integers ranging from 1 to 7.For this study, a question about feedback was adapted becausethis issue did not arise in the implanted patients; thus, thequestion was changed to assess the volume of the speech pro-cessor. For the analyses, the global score was the mean of thescores for all items completed by the patient, and the 20th and80th percentiles gave a range of 4.3 to 5.9 that was selected torepresent patient satisfaction. If the score was below this range,the patient was considered unsatisfied, whereas a score higher than this range was very satisfied.The audiometric thresholds and percentages of correct res- ponsesonspeechperceptiontests foreachconditionare presentedas means and standard deviations (SDs), as well as medians and25th and 75th percentiles (P25% and P75%, respectively). Anonprobability sample was performed because of the explora-tory nature of this study.Considering the small sample size, ordinal nature and asym-metric distributions of some variables, a nonparametric Wilcoxontest for paired data was used to compare groups. A sensitivityanalysis using the parametric  t  -test was performed for all com- TABLE 4.  Choice of converted or upgraded maps and frequency table for each patient  Patient Chosen mapMap with better visualanalogue scaleMap with best  performanceSpectra frequencytable (Hz)Freedom frequencytable (Hz)S1 UM UM UM 150  Y  6,730 188  Y  7,938S2 UM UM CM 133  Y  7,008 188  Y  7,938S3 CM CM CM 133  Y  7,008 133  Y  7,008S4 CM CM CM 150  Y  5,744 150  Y  5,744S5 CM CM CM 120  Y  8,658 120  Y  7,938S6 UM Equal CM 150  Y  10,823 188  Y  7,938S7 CM CM CM 120  Y  8,658 120  Y  7,938S8 UM Equal UM 133  Y  7,008 188  Y  7,938S9 CM CM Equal 120  Y  8,658 120  Y  7,938S10 CM Equal CM 120  Y  8,658 120  Y  7,938S11 UM UM Equal 120  Y  8,658 188  Y  7,938S12 CM CM CM 120  Y  8,658 120  Y  7,938S13 CM Equal CM 120  Y  8,658 120  Y  7,938S14 CM CM CM 600  Y  4,600 600  Y  4,600S15 CM CM Equal 120  Y  8,658 120  Y  7,938S16 UM UM CM 120  Y  8,658 188  Y  7,938S17 CM CM CM 120  Y  8,658 120  Y  7,390CM indicates converted map; UM, upgraded map. TABLE 3.  Results of the sound field threshold with both speech processors Spectra map (SM) Best freedom map (P1 or P2)Median (P25%, P75%) Mean (SD) n Median (P25%, P75%) Mean (SD) n  z p  (Wilcoxon)250 Hz 50 (45, 55) 49.4 (13.3) 17 35 (30, 40) 35.0 (9.8) 17 3.529 0.0004500 Hz 55 (45, 55) 52.1 (7.5) 17 40 (35, 40) 37.6 (5.6) 17 3.552 0.00041,000 Hz 45 (45, 50) 47.4 (6.6) 17 30 (25, 35) 29.7 (6.2) 17 3.667 0.00021,500 Hz 45 (40, 50) 46.5 (8.4) 17 30 (25, 35) 30.0 (8.8) 17 3.612 0.00032,000 Hz 50 (40, 55) 47.6 (8.9) 17 30 (25, 35) 31.2 (8.2) 17 3.608 0.00033,000 Hz 50 (45, 55) 49.1 (8.9) 17 35 (25, 40) 33.5 (9.3) 17 3.504 0.00054,000 Hz 55 (45, 55) 53.5 (9.5) 17 45 (30, 45) 38.5 (9.1) 17 3.567 0.00046,000 Hz 55 (45, 65) 58.2 (12.9) 17 40 (30, 50) 41.2 (16.3) 17 3.395 0.00078,000 Hz 60 (50, 75) 63.5 (15.2) 17 80 (70, 90) 74.7 (16.4) 17  j 2.637 0.0084BIAP 48.7 (45, 53.7) 50.1 (6.3) 17 35 (28.7, 38.7) 34.3 (5.9) 17 3.624 0.0003 4 A. T. DE MATOS MAGALHA˜  ES ET AL. Otology & Neurotology, Vol. 00, No. 00, 2013
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks