Auditory Brainstem Response Detection Research Articles

Optimising weighted averaging for auditory brainstem response detection

The auditory brainstem response (ABR) is a clinical test used to evaluate hearing objectively. The aim of this study was to optimise weighted averaging for both residual noise reduction and also for objective ABR detection using the Fmp statistical test. Analyses were performed using no-stimulus EEG background activity recorded from 15 participants and simulated “response present” data (4,602 ensembles in total). Different approaches for estimating the variance of the noise within each block were compared, as was the effect of the number of recording epochs in each block when calculating and applying the weights. The “VAR Whole Block” method was found to be more effective than the “VAR MP” method at estimating the noise level, especially for smaller block sizes (2–10 epochs). Caution should be exerted when selecting recording parameters for use with weighted averaging as an inflation in the “response absent” Fmp statistic was observed using small block sizes (relative to unweighted averaging); this may be due to a bias in the Fmp statistic observed as a result of the combined effects of the finite Fmp analysis window length and the high-pass filter setting. Optimised weighted averaging was effective in reducing the mean residual noise level in the averaged waveform, leading to improved ABR detection. Further work is required to optimise the Fmp analysis window length, recording settings, and weighted averaging parameters in combination, using a large clinical dataset.

Head and neck radiotherapy causes significant disruptions of cochlear ribbon synapses and consequent sensorineural hearing loss

PurposeTo investigate the possible effects of head and neck radiotherapy on hearing function in mice. MethodAdult C57BL/6J mice were irradiated to the head and neck once with cobalt-60 rays at doses of 10 Gy or 20 Gy. Hearing function was estimated by the detection of auditory brainstem response (ABR) thresholds and the suprathreshold function of cochlear was indicated by the peak amplitudes and latencies of wave I. The mice were tested on days 1, 7, 14, and 21 after radiation treatment, and untreated mice in littermates served as controls. The cochlear pre-synaptic ribbons were labeled using an anti-RIBEYE/CtBP2 antibody, and the synaptic vesicle membrane was traced using anti-vesicular glutamate transporter 3 (VGLUT-3) antibody. The number and size of the pre-synaptic ribbons were counted along the cochlear axis from the apex to the base. The expression of VGLUT-3 was measured by the intensity of immunofluorescence. Hematoxylin and eosin (H&E) staining was also performed to evaluate the structural changes in the cochlea. ResultsCompared with the controls, mice treated with 10 Gy and 20 Gy doses on days 1, 7, 14, and 21 were found to have significant disruptions in ABR thresholds and amplitudes (p < 0.05). Moreover, mice in the 20 Gy group, compared with the 10 Gy group, showed greater hearing loss and suprathreshold deficits (p < 0.05). Quantitative analysis revealed a decrease in the number and size of CtBP2-positive puncta in both the 10 Gy and 20 Gy groups compared with the controls (p < 0.05); in the 20 Gy group, the number and size of CtBP2-positive puncta were less than those in the 10 Gy group (p < 0.05). We observed a significant disruption in the expression of VGLUT-3 in the group treated with 20 Gy. However, compared with the control group, both immunofluorescence and H&E staining revealed no significant changes in the number of hair cells or the array for the 10 or 20 Gy treatments (p > 0.05). ConclusionRadiation therapy targeting the head and neck can cause sensorineural hearing loss via disruption specific to the cochlear ribbon synapses. To our knowledge, this is the first study to demonstrate that cochlear ribbon synapses may be a subcellular target of radiation-induced hearing loss.

Auditory Brainstem Response Detection Using Machine Learning: A Comparison With Statistical Detection Methods.

The primary objective of this study was to train and test machine learning algorithms to be able to detect accurately whether EEG data contains an auditory brainstem response (ABR) or not and recommend suitable machine learning methods. In addition, the performance of the best machine learning algorithm was compared with that of prominent statistical detection methods. Four machine learning algorithms were trained and evaluated using nested k-fold cross-validation: a random forest, a convolutional long short-term memory network, a stacked ensemble, and a multilayer perceptron. The best method was evaluated on a separate test set and compared with conventional detection methods: Fsp, Fmp, q-sample uniform scores test, and Hotelling's T2 test. The models were trained and tested on simulated data that were generated based on recorded ABRs collected from 12 normal-hearing participants and no-stimulus EEG data from 15 participants. Simulation allowed the ground truth of the data ("response present" or "response absent") to be known. The sensitivity of the best machine learning algorithm, a stacked ensemble, was significantly greater than that of the conventional detection methods evaluated. The stacked ensemble, evaluated using a bootstrap approach, consistently achieved a high and stable level of specificity across ensemble sizes. The stacked ensemble model presented was more effective than conventional statistical ABR detection methods and the alternative machine learning approaches tested. The stacked ensemble detection method may have potential both in automated ABR screening devices as well as in evoked potential software, assisting clinicians in making decisions regarding a patient's ABR threshold. Further assessment of the model's generalizability using a large cohort of subject recorded data, including participants of different ages and hearing status, is a recommended next step.

A group sequential test for ABR detection

Objective: To detect the auditory brainstem response (ABR) automatically using an innovative sequentially applied Hotelling’s T 2 test, with the overall goal of optimising test time whilst controlling the false-positive rate (FPR).Design: The stage-wise critical decision boundaries for accepting or rejecting the null hypothesis were found using a new approach called the Convolutional Group Sequential Test (CGST). Specificity, sensitivity, and test time were evaluated using simulations and subject recorded data.Study sample: Data consists of click-evoked ABR threshold series from 12 normal hearing adults, and recordings of EEG background activity from 17 normal hearing adults.Results: Reductions in mean test time of up to 40–45% were observed for the sequential test, relative to a conventional “single shot” test where the statistical test is applied to the data just once. To obtain these results, it will occasionally be necessary to run the test to a higher number of stimuli, i.e. the maximum test time needs to be increased.Conclusions: The CGST can be used to control the specificity of a sequentially applied ABR detection method. Doing so can reduce test time, relative to the “single shot” test, when considered across a cohort of test subjects.

Protocol for Rapid, Accurate, Electrophysiologic, Auditory Assessment of Infants and Toddlers

Background: Audiologists often lack confidence in results produced by current protocols for diagnosticelectrophysiologic testing of infants. This leads to repeat testing appointments and slow protocols whichextend the time needed to complete the testing and consequently delay fitting of amplification. A recentpublication (Sininger et al, 2018) has shown how new technologies can be applied to electrophysiologictesting systems to improve confidence in results and allow faster test protocols. Average test times forcomplete audiogram predictions when using new technologies and protocols were found to be just over32 minutes using auditory brainstem response (ABR) and just under 20 minutes using auditory steadystateresponse (ASSR) technology.&lt;br /&gt;Purpose: The purpose of this manuscript is to provide details of expedited test protocols for infant andtoddler diagnostic electrophysiologic testing.&lt;br /&gt;Summary: Several new technologies and their role in test speed and confidence are described includingCE-Chirp stimuli, automated detection of ABRs using a technique called FMP, Bayesian weighting which isan alternative to standard artifact rejection and Next-Generation ASSR with improved response detectionand chirp stimuli. The test protocol has the following features: (1) preliminary testing includes impedancemeasures and otoacoustic emissions, (2) starting test levels are based on Broad-Band CE-Chirp thresholdsin each ear, (3) ABRs or ASSRs are considered present based on automated detection rather thanon replication of responses, (4) number of test levels is minimized, (5) ASSR generally evaluates fourfrequencies in each ear simultaneously with flexibility to change all test levels independently.&lt;br /&gt;Conclusions: Combining new technologies with common-sense strategies has been shown to substantiallyreduce test times for predicting audiometric thresholds in infants and toddlers (Sininger et al, 2018).Details and rationales for changing test strategies and protocols are given and case examples are used toillustrate.

Evaluation of waveform, latency and amplitude values of chirp ABR in newborns

ObjectivesThere has been a rapid growth in using chirp stimuli for evoking auditory brainstem response (ABR). This study aims to investigate the waveform morphology, latency, and amplitude values in newborns and compare them with those evoked by the click in order to determine whether chirp-evoked responses are easier to detect at near-threshold values, as has been proposed, and to assess the usefulness of the chirp stimulus in improving ABR detection thresholds in this population. MethodsChirp- and click-evoked ABR were measured in 96 normal-hearing newborns (52 males and 44 females) using the MB11 BERAphone®. Only children who passed the hearing screening and were not older than 5 days were included in this study. Two stimulation levels were used: 40dB nHL and 60dB nHL. The stimulation rate was 20.3s–1. The recorded ABR were analyzed based on visual identification of waves I, III, and V, and their amplitude and latency values. ResultsChirp-evoked responses showed clearly larger wave amplitudes compared to those evoked by click stimulus. The gain in amplitude for wave V was 77% for 60dB nHL and 100% for 40dB nHL, for wave III the gain was 22% at 60dB nHL and 46% at 40dB nHL and for wave I the gain was 64% and 100%, respectively. Generally, the gain in amplitude associated with the chirp was greater at the lower stimulus level of 40dB nHL. The wave's latencies were shorter in response to the chirp stimulus. Interpeak intervals differed between click-evoked and chirp-evoked ABR, slightly in the I–III interval and significantly in the I–V interval. ConclusionsChirp-evoked potentials are distinguished from click-evoked potentials by their significantly larger ABR amplitudes. Therefore they are expected to yield more reliable results and shorter measurement times. It should be noted that, for diagnostic purposes, the typical peak latencies of chirp-evoked ABR are clearly different to those of clicks.

ABR in the diagnosis of vestibular schwannomas: A meta-analysis

PurposeThe aim of this study is to rigorously evaluate the role of auditory brainstem response (ABR) testing in the diagnosis of vestibular schwannomas (VS). Materials and methodsSearches were conducted in multiple online databases, supplemented by hand searches. From the studies chosen for final inclusion, relevant data were extracted and meta-analysis of pooled data was performed. Results623 studies were identified from which 43 met inclusion criteria for analysis (1978 to 2009) including 3314 patients. Pooled sensitivity for ABR detection of vestibular schwannomas was 93.4% (95% CI 92.6–94.3, P=0.0000). For tumors less than 1cm (8 studies, 176 patients) sensitivity was 85.8% (95% CI 80.6–90.1, P=0.0116). For tumors greater than 1cm (6 studies, 251 patients) pooled sensitivity was 95.6% (95% CI 93.1–98.2, P=0.0660). Sensitivity of ABR to detect extracanalicular tumors was higher than for intracanalicular tumors, though pooled data were not statistically valid. Pooled specificity (8 studies, 2432 patients) was 82.0% (95% CI 80.5–83.6, P=0.0000). ConclusionsAlthough MRI remains the gold standard, emerging trends towards more conservative management coupled with limited financial resources may prompt many clinicians to review the role of ABR testing in screening for retrocochlear pathology. In light of the high sensitivity and specificity of ABR testing for VS, we strongly urge its reconsideration as a useful diagnostic tool for patients with clinically suspected VS.

ABR-based newborn hearing screening with MB11 BERAphone® using an optimized chirp for acoustical stimulation

ObjectiveAt our center, the Maico MB11 BERAphone® device is used for newborn hearing screening based on Auditory Brainstem Responses (ABR). In 2006, an optimized chirp stimulus was implemented in the device to increase the reliability and quality of the screening method. In 2002, an automated response detection algorithm had been implemented. This study analyzes the screening results using the MB11 BERAphone® device with the implemented chirp stimulus and automated response detection method. MethodsThe data presented were collected in the well-baby nursery as part of the newborn hearing screening program following a two stage screening protocol. To focus the study on the typical routine screening, data from at-risk babies were not included. Overall, data from 6866 babies (3604 males and 3262 females) screened from March 2006 to April 2011 were analyzed in this study. ResultsOut of the 6866 babies screened, 6607 passed bilaterally prior to hospital discharge (defined as 1st stage in this hearing screening program). Therefore, the pre-discharge pass rate of the hearing screening with the MB11 BERAphone® device was 96.2%. The resulting referral rate was 3.8%. The median test time per ear (excluding time for preparation and data reporting) was 28s with a range of 15–112s (5–95th percentile).The number of infants referred for 2nd stage, post-discharge re-screening was 259. Of this group, 71 passed bilaterally and 188 failed the re-screening in one or both ears. Therefore, including both the pre-discharge and post-discharge screening results, the bilateral pass rate was 97.3% and 2.7% were referred for diagnostic evaluation.Diagnostic testing was performed on all of the 188 infants who were referred. Results showed that 47 of these babies had hearing loss. This equates to a positive predictive value for a refer result of 25%. The observed prevalence of hearing impairment in our population was 0.684%.Diagnostic results for 141 of the referred newborns proved that they had normal hearing. That is, 141 out of 6866 newborns had a false-positive result with the MB11 BERAphone® screening. Therefore, the device had a specificity of 97.9%. During the time period of this study, no baby who passed the hearing screening was later found to have hearing impairment, suggesting a sensitivity of 100%. However, due to the limited number of newborns and the short time period after screening for the more recently screened babies, a reliable estimate of the sensitivity cannot be yet made from our program statistics. The theoretical sensitivity of the MB 11 device is 99.9%. ConclusionsThe Maico MB11 BERAphone® is a reliable device for auditory brainstem response newborn hearing screening and it provides results within a very short time. Due to the implemented method for automatic detection of ABR, the use of the device does not require an experienced examiner, allowing it to be used by trained technicians in locations outside the department of audiology such as the well-baby nursery.

Click-evoked ABR at high stimulus repetition rates for neonatal hearing screening: Potenciales auditivos evocados (ABR) por clicks a tasas altas de estimulación para la identificación de problemas auditivos en neonatos

A new, fast screening algorithm based on auditory brainstem response (ABR) recorded at a high click repetition rate is proposed. Response detection is carried out in the frequency domain by a statistical test procedure which includes the fundamental frequency and the harmonics below 800 Hz. First, the method was tested in 25 young adults. ABRs were recorded in the repetition rate range 20/s to 400/s. With a mean response detection time of 31 s, a click repetition rate of 140/s was found to be the optimum rate among the adult group. The method was then tested using a group of 114 neonates in whom the repetition rate range 60/s to 200/s was examined. At the repetition rate 90/s, which was found to be the optimum rate in neonates, the mean detection time was 24.6 s. In addition to the fast ABR detection, the proposed screening algorithm also allows simultaneous hearing screening of both ears using a one-channel data recording.Se propone un algoritmo nuevo y rápido de identificación basado en las respuestas auditivas de tallo cerebral (ABR) registradas con una tasa alta de estimulación con clicks. La detección de las respuestas se lleva a cabo en el campo fre-cuencial por medio de una prueba estadistica que incluye la frecuencia fundamental y los armónicos inferiores a 800 Hz. El método fue inicialmente probado en 25 adultos jóvenes. Las ABRs se registraron con un rango de tasas de estimulación de 20/s a 400/s. Se encontró que con una tiempo promedio de detección de la respuesta de 31 s., la tasa óptima de estimulación con clicks, para el grupo de los adultos era de 140/s. Posteriormente el método fue probado en un grupo de 114 neonatos en quienes la tasa de estimulación varió de 60/s. a 200/s. El tiempo promedio de detección fue de 24.6 s., cuando se utilizó una tasa de estimulación de 90/s., considerada como la optima para neonatos. Además de la rápida detección de ABR, el algoritmo de identificación que se propone, pcrmite también la identificación auditiva simul-tánea en ambos oídos con el uso de un sistema de registro de datos de un canal.

Latency estimation of auditory brainstem response by neural networks

In the clinical application of auditory brainstem responses (ABRs), the latencies of five to seven main peaks are extremely important parameters for diagnosis. In practice, the latencies have mainly been done by manual measurement so far. In recent years, some new techniques have been developed involving automatic computer recognition. Computer recognition is difficult, however, since some peaks are complicated and vary a lot individually. In this paper, we introduce an artificial neural network method for ABR research. The detection of ABR is performed by using artificial neural networks. A proper bandpass filter is designed for peak extraction. Moreover, a new approach to estimate the latencies of the peaks by artificial neural networks is presented. The neural networks are studied in relation to the selection of model, number of layers and number of neurons in each hidden layer. Experimental results are described showing that artificial neural networks are a promising method in the study of ABR.

Computer analysis of auditory brainstem responses by using advanced pattern recognition

Automatic peak detection of auditory brainstem responses is a multistage process. The main task is the detection of the pedestal peak, which is a broad peak filtered out from the auditory brainstem signal. Special validation stages have been used to determine whether a real response is present, and if so, peak V is detected. The system reaches 90% accuracy with random clinical material.