Standardization of Three Familiar Sound Recognition Tests in Hearing and Deaf Adult Populations.

OBJECTIVE: Recognition of familiar noises is crucial for understanding and reacting appropriately to our auditory environment. Its improvement is one of the benefits expected after cochlear implantation. The aim of this study was to standardize three environmental sounds noise recognition tests and to illustrate their application to a population of deaf adults with cochlear implants. METHOD: Norms were established on a sample of 126 normal-hearing adults divided into 6 age groups. Three familiar sound recognition tests were used: 1) the Blue Mouse "First Familiar Sounds" (BM), 2) the UCL-IRSA test (TI), and 3) the Bernadette Piérart Familiar Sounds Test (TBF). These tests were also administered to 61 implanted deaf ears. RESULTS: We observed a significant effect of age on the accuracy scores of the TI and TBF tests for the hearing group and on the time scores of the TI and BM tests. Overall, the performance of the deaf participants was poorer and more variable than that of the hearing participants. CONCLUSION: We have three tests that can be used in practice to measure the performance of deaf people (with cochlear implants) at different stages of their pre- and post-implant rehabilitation.

Effect of a CI Programming Fitting Tool with Artificial Intelligence in Experienced Cochlear Implant Patients.

OBJECTIVE: Cochlear implants (CIs) are the treatment of choice for patients with severe to profound hearing loss. The hearing results, however, considerably vary across patients. This may partly be due to variability in the CI fitting. We investigated the effect of FOX, a software tool to program CIs using artificial intelligence (AI), on hearing outcomes. METHODS: Forty-seven experienced CI patients who came to our tertiary CI center for their annual follow-up between 2017 and 2020 were recruited for this study. They received a new CI map created by the AI software tool. CI parameters and auditory outcomes obtained with this new map were compared with those of the initial manual map after 15 days of take-home experience. Within-patient differences were assessed. At the end of the study, the patients were offered a choice to continue using the AI map or to revert to their old manual map. RESULTS: Several auditory outcomes improved with the AI map, namely, pure tone audiometric threshold at 6,000 Hz (median improvement 10 dB, range = -20 to 50 dB, Z = -2.608, p = 0.008), phonemic discrimination scores (median improvement 10%, range = 0% to 30%, Z = -4.061, p = 0.001), and soft-intensity (median improvement of 10%, range = -20% to 90%, Z = -4.412, p < 0.001) to normal-intensity (median improvement of 10%, range = -30% to 60%, Z = -3.35, p < 0.001) speech audiometric scores. CONCLUSION: The AI-assisted CI mapping model as a potential assistive tool may improve audiological outcomes for experienced CI patients, including high-frequency pure tone audiometry and audiometric speech scores at low and normal presentation levels.Clinical trial registration: NCT03700268.

Prospective Comparison Between Manual and Computer-Assisted (FOX) Cochlear Implant Fitting in Newly Implanted Patients.

OBJECTIVE: A prospective, longitudinal, randomized controlled trial with an original crossover design for 1 year was conducted to compare manual fitting to artificial intelligence-based fitting in newly implanted patients. DESIGN: Twenty-four patients who received their first cochlear implant (CI) were randomly assigned to the manual or Fitting to Outcome eXpert (FOX) arm; they followed the corresponding fitting procedures for 1 year. After 1 year, each patient was switched to another arm. The number of fittings, auditory outcomes (pure-tone thresholds, loudness scaling curves, spectral discrimination scores, bisyllabic word recognition in quiet and noise, and speech tracking), fitting session questionnaire, and CI parameters (T level, C level, Threshold Sound Pressure Level (T-SPL), Comfortable Sound Pressure Level (C-SPL), and loudness growth value) were compared between the two groups. Differences between the two groups were analyzed using the Mann-Whitney test, and Holm corrections were applied for multiple statistical tests. At the end of the crossover session, patients were offered the choice to continue with their old or new map. RESULTS: As early as 3 mo postactivation, the FOX group showed less variability and significantly better speech intelligibility in quiet conditions at 40 and 55 dB SPL and noise ( p < 0.05) with median phoneme scores of 50%, 70%, and 50% at 55, 70, and 85 dB SPL compared with 45%, 50%, and 40%, respectively. This group showed better results at 12 mo postactivation ( p < 0.05). In the manual group, 100% of the patients decided to keep the new FOX map, and 82% performed better with the FOX map. In the FOX group, 63% of the patients decided to keep the manual map, although the measurable outcome had not improved. In this group, participants reported to prefer the manual map because it felt more comfortable, even if the FOX map gave better measured outcome. CONCLUSION: Although the study size remains relatively small, the AI-FOX approach was equivalent to or even outperformed the manual approach in hearing performance, comfort, and resources. Furthermore, FOX is a tool capable of continuous improvement by comparing its predictions with observed results and is continuously learning from clinicians' practice, which is why this technology promises major advances in the future.

Variability of fitting parameters across cochlear implant centres.

OBJECTIVE: As a follow-up to the studies by Vaerenberg et al. (Sci World J 501738:1-12, 2014) and Browning et al. (Cochlear Implant Int 21(3):1-13, 2020), who used questionnaires, we determined whether there are between-centre variations in the fitting of cochlear implants by analysing the methodology, fitting parameters and hearing results of patients from four centres with real data. The purpose of this study is to highlight the lack of streamlined mapping guides and outcome measures with respect to cochlear implant (CI) fittings. METHODS: A retrospective study with ninety-seven post-lingual adults with a nucleus cochlear implant placed between 2003 and 2013 was included to ensure at least 5 years of follow-up. The studied data were as follows: the methodology, including the fitter's professional background, the method of activation, the sequence of fitting sessions, the objectives measures and hearing outcomes; and the fitting parameters, including the speech processors, programming strategy, stimulation mode, T and C levels, T-SPL and C-SPL, maxima, pulse width, loudness growth and hearing results. RESULTS: This investigation highlights some common practices across professionals and CI centres: the activation of a CI is behavioural; impedances are systematically measured at each fitting; and some parameters are rarely modified. However, there are also differences, either between centres, such as the sequences of fitting sessions (p < 0.05) or their approach to spectral bands (p < 0.05), or even within centres, such as the policy regarding T and C levels at high frequencies compared to those at low and mid-frequencies. CONCLUSION: There are important variations between and within centres that reflect a lack of CI-related policies and outcome measures in the fitting of CI. CLINICAL TRIALS REGISTRY: NCT03700268.

From manual to artificial intelligence fitting: Two cochlear implant case studies.

Objective: To assess whether CI programming by means of a software application using artificial intelligence (AI), FOX®, may improve cochlear implant (CI) performance. Patients: Two adult CI recipients who had mixed auditory results with their manual fitting were selected for an AI-assisted fitting. Even after 17 months CI experience and 19 manual fitting sessions, the first subject hadn't developed open set word recognition. The second subject, after 9 months of manual fitting, had developed good open set word recognition, but his scores remained poor at soft and loud presentation levels. Main outcome measure(s): Cochlear implant fitting parameters, pure tone thresholds, bisyllabic word recognition, phonemic discrimination scores and loudness scaling curves. Results: For subject 1, a first approach trying to optimize the home maps by means of AI-proposed adaptations was not successful whereas a second approach based on the use of Automaps (an AI approach based on universal, i.e. population based group statistics) during 3 months allowed the development of open set word recognition. For subject 2, the word recognition scores improved at soft and loud intensities with the AI suggestions. The AI-suggested modifications seem to be atypical. Conclusions: The two case studies illustrate that adults implanted with manual CI fitting may experience an improvement in their auditory results with AI-assisted fitting.