Resonance R140 - Beta test 2019

• 

Details

Parent Category: OAE Hardware

Category: OAE Equipment Models

Created on Tuesday, 17 September 2019 11:03

Last Updated on Sunday, 27 September 2020 17:54

Written by Editorial Board

Hits: 3859

EDITOR'S NOTE : We had the chance to evaluate an interesting 4th generation AOAE device prior to its official release the R140 by Resonance (Italy). On the average, the device performed well but it required some additional optimization prior to the official release. In late 2019 the manufacturer had promised to provide the device for another set of measurements, due to the Covid situation this phase was never implemented.

 

Technical Report

Evaluation of the R140 Resonance OAE device

 

Ferrara, April 2019

INTRODUCTION

The objective of this pilot study was to evaluate the clinical performance of the Resonance R140 device in infants and adults. The touch-screen Accuscreen screening device was used as the gold OAE standard.

 

MATERIALS and METHODS

 

Testing Period:

The pilot project started in September 2018 and ended in December 2018. The project was coordinated by the clinic of Audiology and ENT, in conjunction with the clinic of Pediatrics, of Ferrara University.

 

Subjects

1. 50 adult volunteers participated in the study and were subdivided in 3 sub-groups according to their age: GR1: 8-12; GR2: 21-23; GR3: 24 -64. All subjects were assessed with pure tone audiometry in order to classify their hearing into the standard hearing loss classes (normal, border-line normal, mild, moderate hearing loss etc.). The subjects were only assessed with R140 DPOAE and TEOAE protocols.
2. 70 neonates were assessed during the second day of life, having an age > 24 hrs. The testing time was dictated by data in the literature (4,5,6) suggesting that after 24 hrs, many sound transmissive factors are no longer present in the external canal or in the middle ear, thus facilitating the recording of the OAE response. The subjects were assessed with DPOAE and TEOAE protocols (R140) and by TEOAE protocols (Accuscreen, Otometrics, Italy). The performance of Accuscreen was considered as the clinical standard.

 

OAE Protocols

For the R140, the adult and neonatal subjects were assessed with 3 OAE protocols

(i)            By a 70-70 dB SPL DPOAE protocol testing the frequency range from 2.0 to 4.0 kHz (3 testing points). If the response was not evaluated as a PASS in these 3 points, additional higher frequencies were also tested (i.e. > 5.6 kHz).

(ii)          By a TEOAE protocol (TEOAE clinical) evaluating the signal to Noise ratio (S/R) of the response, at the frequency range from 1.0 to 4.0 kHz (5 frequency bands).

(iii)         By a TEOAE protocol evaluating the TEOAE response in the 1.0 – 4.0 kHz range, by a statistical algorithm (TEOAE screening). The latter is similar to the statistical algorithm the Accuscreen device uses.

 

For the Accuscreen testing the default TEOAE protocol was used only on the neonatal subjects. Additional information on the protocol can be found in the the white paper by Dr. Armin Giebel on the foundations of the statistical algorithm employed in the Natus – Otometrics family of hearing screening devices.

       In terms of timing (first, second, third) the 4 protocols were randomly executed.

Statistical Analyses

The data were analyzed under the assumption of a normal distribution of the tested variables. No correction factors were introduced into the data sets. Various differences among the R140 TEOAE and DPOAE responses were evaluated by standard 2-tailed t-tests. A p < 0.05 was considered significant.

 

RESULTS from the ADULT groups

 

From the 3 tested protocols the one which generated the largest number of PASSes was the DPOAE protocol. Tables 1, 2, 3 show the performance of the protocols per subgroup (i.e. GR1, GR2, GR3). The analytical data on the distribution of the response amplitude (TEOAE-clinical and DPOAE) are included in the Appendix.

	PASS	REFER	NC
DPOAE	5	0	0
TEOAE-clinical	3	1	1
TEOAE-Screening	4	1	0

Table 1: Screening evaluation in the GR1 group (5 subjects): NC= Non Conducted

In the GR1 group the best performance was seen in the DPOAE protocol . Both TEOAE protocols presented REFER  cases and only the TEOAE -clinical protocol generated a NC status (the recording could not be terminated).

 

	PASS	REFER	NC
DPOAE	4	0	0
TEOAE-clinico	4	0	0
TEOAE-SCR	4	0	0

Table2: Screening evaluation in the GR2 (4 subjects): NC = Non Conducted.

In the GR2 group all protocols generated the same PASS outcomes. It is worthwhile to mention that the DPOAE responses were characterized by excellent S/N ratios , exceeding in some cases 15 dB .

 

 

	PASS	REFER	NC
DPOAE	40	1	0
TEOAE-clinical	18	22	1
TEOAE-Screening	7	21	14

Table 3: Screening evaluation in the GR3 group (41 subjects): NC = Non Conducted.

 

(i)            Table 3 shows that the DPOAE protocol generated the largest number of PASS outcomes (40/41). The outcome difference (yield) between the OAE protocols was found to be  large, specially between the DPOAE and the TEOAE-screening protocol yields (see comment below).

(ii)          For the GR3 adult group, the TEOAE-Screening protocol presented many inconclusive cases (14/41), probably due to the fact that the algorithm was calibrated for neonatal and not adult ears.

(iii)         The performance of the TEOAE protocols can be improved by optimizing better the screening protocol parameters. For a number of cases a REFER outcome could be transformed into a PASS considering less frequency points (3 out of 4) or different (lower) values of the classifying criteria.

(iv)         The DPOAE and TEOAE-clinical protocols captured different aspects of the OAE response. The data from the frequency distribution (See the Appendix Section 1) are different. Table 4 presents a signal comparison between protocols at 3.0 kHz, where the mean response values are significantly different. A partial explanation of the observed difference is the fact that the DPOAE protocol is referenced to 3.3 and not 3.0 kHz.

 

	TEOAE SN3	DP SN3_3
Mean	3,018367347	10,772
Variance	29,08319728	415,599033
Observations	49	50
Hypothesized Mean Difference	0
df	56
t Stat	-2,598218643
P(T<=t) one-tail	0,005976819
t Critical one-tail	1,672522303
P(T<=t) two-tail	0,011953638
t Critical two-tail	2,003240719

 Table  4:t-test results from the GR3 TEOAE e DPOAE comparison at 3 kHz . The test assumed unequal variances .

 

RESULTS FROM THE NEONATAL GROUP

 

The performance of the 3 R140 protocols vs Accuscreen is reported in Table 5. The TEOAE protocols presented a high number of unconcluded recordings (36 and 41 out of 70). The best performer, as in the case of the adult subjects, was the DPOAE protocol.

 

	PASS	REFER	NC
Accuscreen	62 (88,5%)	8	0
DPOAE	57 (93,4%)	4	9
TEOAE-clinical	17 (50%)	17	36
TEOAE-Screening	20 (68,9%)	9	41

Table  5: Clinical performance of the 4 tested protocols NC = NON Conducted

 

1. Although the PASS outcome of Accuscreen, in terms of subjects, is higher than the DPOAE protocol (62 vs 57), the latter generated less REFER cases and therefore presented a higher outcome ratio (93,4%) than the Accuscreen.
2. There is a considerable difference between the outcomes of the two statistical TEOAE protocols (88,5% vs 68,9%).  Interestingly, the number of REFERS between the two is almost identical (8 vs 9). Overall the Accuscreen performs better, but the R140 protocol is probably penalized by the high number of inconclusive recordings. Despite the fact that the timing of the execution of the protocols was randomized, there are no data on the amount of ambient noise during the execution of the R140 TEOAE-Screening protocol, thus it is unknown if the number of inconclusive recordings is related to high levels of acoustic noise.
3. As for the adult subjects, the R140 OAE protocols captured different aspects of the OAE response. Table 6 shows the DPOAE-TEOAE S/N differences at 3 kHz. 

 

Table  6: T-test results from the TEOAE e DPOAE comparison at 3 kHz, in the neonatal group.  The test assumed unequal variances.

As in the case of adult subjects, part of the observed signal differences might be caused by the fact that the DPOAEs are referenced to 3.3 kHz. An additional factor is, that the TEOAE protocol has generated significantly less responses (35 vs 61) than the DPOAE one.