Search

Tinnitus and age-related hearing difficulties can occur while hearing sensitivity remains normal. Because the exact mechanisms underlying these functional deficits in clinically normal-hearing listeners are poorly understood, it remains difficult to select a suitable treatment. To address this knowledge gap, we studied how age and tinnitus factors (separately and in combination) affect speech intelligibility in quiet and noise. We specifically focused on low (1.65 kHz) hearing mechanisms associated with temporal fine-structure (TFS) and temporal envelope (TENV) coding, and complemented our behavioural assays with physiological and early-markers of sensorineural hearing loss (SNHL). Older listeners with normal audiograms (upto 4 kHz) had reduced extended high-frequency hearing sensitivity, reduced physiological markers of temporal envelope coding and compromised auditory-nerve integrity compared to young normal-hearing controls. These age-related declines were similar across tinnitus- and non-tinnitus patients, showing a stronger impact of age than tinnitus on the considered SNHL markers. High-frequency speech-intelligibility was similar in tinnitus and control subjects, and we observed a perceptual low-frequency speech-in-noise (but not speech-in-quiet) benefit in older tinnitus subjects. Further insight into the hearing mechanisms that improved low-frequency speech-in-noise coding in older tinnitus subjects is necessary, and could further the development of individualised treatments., Funding:UGent BOF-IOP: EarDiMon

Current hearing aids mostly provide sound amplification fittings based on individual hearing thresholds or perceived loudness, even though it is known that sensorineural hearing damage is functionally complex, and requires different treatment strategies. To meet this demand, we propose an optimisation framework for the design of individualised hearingaid signal processing based on simulated (hearing-impaired) auditory-nerve responses. The framework is fully differentiable, thus the backpropagation algorithm can be used to train DNN-based hearing-aid models that optimally process sound to restore hearing in impaired cochleae. The auditory models within the framework can be tuned to the precise hearing-loss profile of a listener to yield trully individualised restoration strategies. Our simulations show that the trained hearing-aid models were able to enhance the auditory-nerve responses of hearing-impaired cochleae, and this provides a promising outlook for embedding our framework within future hearing aids and augmented-hearing applications.

Environmental epidemiology 6(1), e183 (2022). doi:10.1097/EE9.0000000000000183, Published by Wolters Kluwer Health, Alphen aan den Rijn

Purpose: Speech-in-noise tests and suprathreshold auditory evoked potentials are promising biomarkers to diagnose cochlear synaptopathy (CS) in humans. This study investigated whether these biomarkers changed after recreational noise exposure.Method: The baseline auditory status of 19 normal-hearing young adults was analyzed using questionnaires, pure-tone audiometry, speech audiometry, and auditory evoked potentials. Nineteen subjects attended a music festival and completed the same tests again at Day 1, Day 3, and Day 5 after the music festival.Results: No significant relations were found between lifetime noise-exposure history and the hearing tests. Changes in biomarkers from the first session to the follow-up sessions were nonsignificant, except for speech audiometry, which showed a significant learning effect (performance improvement).Conclusions: Despite the individual variability in prefestival biomarkers, we did not observe changes related to the noise-exposure dose caused by the attended event. This can indicate the absence of noise exposure–driven CS in the study cohort, or reflect that biomarkers were not sensitive enough to detect mild CS. Future research should include a more diverse study cohort, dosimetry, and results from test–retest reliability studies to provide more insight into the relationship between recreational noise exposure and CS.Supplemental MaterialThis section provides additional information on the relation between Laeq,life and EHF audiometric thresholds of different (Supplemental Material S1) and averaged frequencies (Supplemental Material S2), the variation of speech reception thresholds of different conditions in relation to Laeq,life (Supplemental Materials S3 and S4) and correlation between Laeq,festival and shifted amount of different AEP markers as well as EHF audiogram from session 1 to session 2 (Supplemental Material S6). Furthermore, the amplitude and latency shifts of ABR wave I and V from session 1 to session 2 are reported in Supplemental Material S5. The shifted amounts are specified separately for groups who did or did not experience HRS after attending the music-event.Supplemental Material S1. Spearman’s correlation analysis between Laeq,life and (A) 10 kHz, (B) 12.5 kHz, (C) 14 kHz and (D) 16 kHz EHF thresholds. Correlation coefficients and p-values are indicated in respective panels.Supplemental Material S2. Spearman’s correlation analysis between Laeq,life and averaged EHF audiometric thresholds from 10 to 16 kHz.Supplemental Material S3. Spearman’s correlation analysis between Laeq,life and (A) LP SPiQ, and (B) HP SpiQ. Correlation coefficients and p-values are indicated in respective panels.Supplemental Material S4. Spearman’s correlation analysis between Laeq,life and (A) LP SPiN, (B) HP SPiN and (C) BB SPiN reception thresholds. Correlation coefficients and p-values are indicated in respective panels.Supplemental Material S5. U-values and significance levels of Mann Whitney U test for AEP measurements. Median test shifts from session 1 to session 2, minima and maxima are shown for subjects that have experienced Hearing Related Symptoms (HRS) and those who did not. Significant between-group differences are highlighted with *(p < .05), **(p < .01) or ***(p < .001).Supplemental Material S6. Spearman’s correlation between the Laeq,festival and the shift-amount of different markers from session-1 to session-2: (A) RAM-EFR magnitude, (B) SAM-EFR magnitude, (C) ABR wave-I amplitude, (D) ABR wave-V amplitude, (E) ABR wave-I latency, (F) ABR wave-V latency and (G) EHF threshold shifts. In each panel, a negative shift is an indicator of reduction in the corresponding from session-1 to session-2.Maele, T. V., Keshishzadeh, S., De Poortere, N., Dhooge, I., Keppler, H., & Verhulst, S. (2021). The variability in potential biomarkers for cochlear synaptopathy after recreational noise exposure. Journal of Speech, Language, and Hearing Research. Advance online publication. https://doi.org/10.1044/2021_JSLHR-21-00064

International audience; Biophysically realistic models of the cochlea are based on cascaded transmission-line (TL) models which capture longitudinal coupling, cochlear nonlinearities, as well as the human frequency selectivity. However, these models are slow to compute (in the order of seconds/minutes), explaining why less-accurate model descriptions of cochlear processing (e.g., gammatone, DRNL, MFCC) are still the standard for feature extractors or auditory front-ends. To overcome this gap, we present a hybrid approach in which convolutional neural network (CNN) techniques are combined with computational modelling to yield a real-time model of the human auditory periphery. A CNN was trained on speech corpus material to mimic a state-of-the-art biophysical model that can accurately represent the human cochlea and the ascending auditory pathway. The performance was compared against human data and simulations of the original model using basic stimuli (pure tones, clicks, etc.). Because the original peripheral model can simulate different degrees of sensorineural hearing loss, the normal-hearing CNN model can be adjusted in the same fashion to simulate hearing impairment. The neural-network character of these architectures allows for real-time, parallel and differentiable computations, which can serve in the next generation of hearing-aid and machine-hearing applications. Work supported by European Research Council ERC-StG-678120 (RobSpear)

In classical computational neuroscience, analytical model descriptions are derived from neuronal recordings to mimic the underlying biological system. These neuronal models are typically slow to compute and cannot be integrated within large-scale neuronal simulation frameworks. We present a hybrid, machine-learning and computational-neuroscience approach that transforms analytical models of sensory neurons and synapses into deep-neural-network (DNN) neuronal units with the same biophysical properties. Our DNN-model architecture comprises parallel and differentiable equations that can be used for backpropagation in neuro-engineering applications, and offers a simulation run-time improvement factor of 70 and 280 on CPU or GPU systems respectively. We focussed our development on auditory neurons and synapses, and show that our DNN-model architecture can be extended to a variety of existing analytical models. We describe how our approach for auditory models can be applied to other neuron and synapse types to help accelerate the development of large-scale brain networks and DNN-based treatments of the pathological system., Drakopoulos et al developed a machine-learning and computational-neuroscience approach that transforms analytical models of sensory neurons and synapses into deep-neural-network (DNN) neuronal units with the same biophysical properties. Focusing on auditory neurons and synapses, they showed that their DNN-model architecture could be extended to a variety of existing analytical models and to other neuron and synapse types, thus potentially assisting the development of large-scale brain networks and DNN-based treatments.