Evaluating the Noise in Electrically Evoked Compound Action Potential Measurements in Cochlear Implants

Undurraga, Jaime A.; Carlyon, Robert P.; Wouters, Jan; Wieringen, Astrid Van

doi:10.1109/tbme.2012.2194292

Cited by 11 publications

(8 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Among the automatic ECAP measurement algorithms, AutoNRT ™ can be seen to mimic the approach of “first visual” whereas AutoART follows the method of interpolation. The present FG-SNR algorithm and the previously presented SNR algorithms [ 10 , 11 ] are, in our opinion, closer to the method of “first visual”. Since the FG-SNR algorithm bases the ECAP threshold determination on a sigmoidal fit on binary (ECAP yes/no) classifications instead of fitting a function to the ECAP AGF, the FG-SNR algorithm could theoretically detect the ECAP threshold at a lower stimulus strength than clinicians or algorithms relying partially on the interpolation.…”

Section: Discussionsupporting

confidence: 69%

“…The SNR relates a desired signal (here: ECAP response) to background noise–a high SNR (above an application-specific level) means there is a high level of signal and a low level of background noise, which can be exploited for signal detection. Specifically, SNR-based ECAP threshold estimation can be achieved in one of two ways: (1) estimation can be based on the post-average residual noise and the useful variance [ 10 ] or (2) by a simple comparison of variances calculated for two different time windows within one recording. Hereby, the variance within that part of the recording window that potentially contains an ECAP response is compared to the variance within another part of the recording window that is known not to contain an ECAP response [ 11 ].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Optimized SNR-based ECAP threshold determination is comparable to the judgement of human evaluators

et al. 2021

View full text Add to dashboard Cite

In cochlear implant (CI) users, measurements of electrically evoked compound action potentials (ECAPs) prove the functionality of the neuron-electrode interface. Objective measures, e.g., the ECAP threshold, may serve as a basis for the clinical adjustment of the device for the optimal benefit of the CI user. As for many neural responses, the threshold determination often is based on the subjective assessment of the clinical specialist, whose decision-making process could be aided by autonomous computational algorithms. To that end, we extended the signal-to-noise ratio (SNR) approach for ECAP threshold determination to be applicable for FineGrain (FG) ECAP responses. The new approach takes advantage of two features: the FG stimulation paradigm with its enhanced resolution of recordings, and SNR-based ECAP threshold determination, which allows defining thresholds independently of morphology and with comparably low computational power. Pearson’s correlation coefficient r between the ECAP threshold determined by five experienced evaluators and the threshold determined with the FG-SNR algorithm was in the range of r = 0.78–0.93. Between evaluators, r was in a comparable range of 0.84–0.93. A subset of the parameters of the algorithm was varied to identify the parameters with the highest potential to improve the FG-SNR formalism in the future. The two steps with the strongest influence on the agreement between the threshold estimate of the evaluators and the algorithm were the removal of undesired frequency components (denoising of the response traces) and the exact determination of the two time windows (signal and noise and noise only).”The parameters were linked to the properties of an ECAP response, indicating how to adjust the algorithm for the automatic detection of other neurophysiological responses.

show abstract

Section: Discussionsupporting

confidence: 69%

Section: Introductionmentioning

confidence: 99%

Optimized SNR-based ECAP threshold determination is comparable to the judgement of human evaluators

et al. 2021

View full text Add to dashboard Cite

show abstract

“…The sampling rate of ECAP responses was 56 kHz and the gain was 300. The ECAP responses were processed with a band-pass smoothing filter from 400 to 6000 Hz and were considered nonexisting if the estimated signal-to-noise ratio (SNR) was below 1.7 dB (Undurraga et al 2012). Ideally, ECAP recording should be performed at MCL as in the pitch-ranking test.…”

Section: Ecap: Stimuli and Proceduresmentioning

confidence: 99%

Excitation Patterns of Standard and Steered Partial Tripolar Stimuli in Cochlear Implants

Luo

2015

JARO

View full text Add to dashboard Cite

Current steering in partial tripolar (pTP) mode has been shown to improve pitch perception and spectral resolution with cochlear implants (CIs). In this mode, a fraction (σ) of the main electrode current is returned within the cochlea and steered between the basal and apical flanking electrodes (with a proportion of α and 1 − α, respectively). Pitch generally decreases when α increases from 0 to 1, although the salience of pitch change varies across CI users. This study aimed to identify the mechanism of pitch changes with pTP-mode current steering and the factors contributing to the intersubject variability in pitch-ranking sensitivity. The electrical fields were measured for steered pTP stimuli on the same main electrode with α = 0, 0.5, and 1 in five implanted ears using electrical field imaging (EFI). The related excitation patterns were also measured physiologically using evoked compound action potential (ECAP) and psychophysically using psychophysical forward masking (PFM). Consistent with the pitch-ranking results in this study, the EFI, ECAP, and PFM centroids shifted apically with increasing α. An apical shift was also observed for the PFM peak but not for the EFI or ECAP peak. The pattern width was similar with different α values within a given measure (e.g., EFI, ECAP, or PFM), but the ECAP patterns were broader than the EFI and PFM patterns, possibly because ECAP was measured with smaller σ values than EFI and PFM. The amount of pattern shift with α depended on σ (i.e., the total amount of current used for steering) but was not correlated with the pitchranking sensitivity across subjects. The results revealed that the pitch changes elicited by pTP-mode current steering were not only driven by the shifts of excitation centroid.

show abstract

“…where 𝑴 𝟎𝒂 is the matrix of ECAP amplitudes calculated using sweeps 1-24 to form waveforms for each ECAP, and 𝑴 𝟎𝒃 uses sweeps 25-48. Since halving the number of sweeps increases the measurement noise in the ECAP waveform by a factor of √2 (Underraga, et al, 2012;Stronks, et al, 2019), we incorporated this into the calculation so that 𝜀 0 represents the noise in the 48-sweep waveform ECAP amplitudes that make up 𝑴 𝟎 . To enable a direct comparison between 𝜀 𝑆 and 𝜀 0 , both 𝑴 𝟎𝒂 and 𝑴 𝟎𝒃 are constructed symmetrically in the same manner as descried above for 𝑀 𝑆 matrices, including only the 𝑀 𝑝,𝑚 cells with the same masker and probe conditions as the 𝑴 𝑺 matrices for each participant.…”

Section: Ecap Measurement Assessmentmentioning

confidence: 99%

SpeedCAP: An Efficient Method for Estimating Neural Activation Patterns Using Electrically-Evoked Compound Action-Potentials in Cochlear Implant Users

Garcia¹,

Deeks²,

Goehring³

et al. 2021

Preprint

View full text Add to dashboard Cite

Objectives: Electrically-Evoked Compound Action-Potentials (ECAPs) can be recorded using the electrodes in a cochlear implant (CI) and represent the synchronous responses of the electrically-stimulated auditory-nerve. ECAPs can be obtained using a forward-masking method that measures the neural response to a probe and masker electrode separately and in combination. The Panoramic ECAP (PECAP) method measures ECAPs using multiple combinations of masker and probe electrodes and uses a nonlinear optimization algorithm to estimate current spread from each electrode and neural health along the cochlea. However, the measurement of ECAPs from multiple combinations of electrodes is too time-consuming for use in clinics. This study proposes and evaluates a fast version of the PECAP measurements, SpeedCAP, that minimises recording time by exploiting redundancies between multiple ECAP measures, and that can be applied to methods where multiple ECAPs are required. Design: In the first study, 11 users of Cochlear Limited CIs took part. ECAPs were recorded using the forward-masking artefact-cancellation technique at the most comfortable loudness level (MCL) for every combination of masker and probe electrodes for all active electrodes in the users’ MAPs, as per the standard PECAP recording paradigm. The same current levels and recording parameters were then used to collect ECAPs in the same users with the SpeedCAP method. The ECAP amplitudes were then compared between the two conditions, as were the corresponding estimates of neural health and current spread calculated using the PECAP method described by Garcia et al (2021). The second study measured SpeedCAP intra-operatively in 8 CI patients and with all maskers and probes presented at the same current level to assess feasibility. ECAPs for the subset of conditions where the masker and probe were presented on the same electrode were compared to those obtained using the slower approach leveraged by the standard clinical software. Results: Data collection time was reduced from 45 (PECAP) to 8 (SpeedCAP) minutes. There were no significant differences between normalized root mean squared error (RMSE) repeatability metrics for post-operative PECAP and SpeedCAP data, nor for the RMSEs calculated between PECAP and SpeedCAP data. When between-participant differences were removed, both the neural health (r = 0.73) and current spread (r = 0.65) estimates were significantly correlated (p < 0.0001, df = 218) between SpeedCAP and PECAP conditions across all electrodes. Valid ECAPs were obtained in all patients in the second study, demonstrating intra-operative feasibility of SpeedCAP. No significant differences in RMSEs were detectable between post- and intra-operative ECAP measurements. Conclusions: The improved efficiency of SpeedCAP provides time savings facilitating multi-electrode ECAP recordings in routine clinical practice. The SpeedCAP data collection is sufficiently quick to record intra-operatively, and adds no significant error to the ECAP amplitudes. Such measurements could thereafter be submitted to models such as PECAP to provide patient-specific patterns of neural activation to inform programming of clinical MAPs and/or identify causes of poor performance at the electrode-nerve interface of CI users. The speed and accuracy of these measurements also opens up a wide range of additional research questions to be addressed.

show abstract

Evaluating the Noise in Electrically Evoked Compound Action Potential Measurements in Cochlear Implants

Cited by 11 publications

References 30 publications

Optimized SNR-based ECAP threshold determination is comparable to the judgement of human evaluators

Optimized SNR-based ECAP threshold determination is comparable to the judgement of human evaluators

Excitation Patterns of Standard and Steered Partial Tripolar Stimuli in Cochlear Implants

SpeedCAP: An Efficient Method for Estimating Neural Activation Patterns Using Electrically-Evoked Compound Action-Potentials in Cochlear Implant Users

Contact Info

Product

Resources

About