The Audioscan: a high frequency resolution audiometric technique and its clinical applications

REVIEW

The Audioscan: a high frequency resolution audiometrictechnique and its clinical applications

F. ZHAO,� D. STEPHENSy & C. MEYER-BISCHz�Cardiff University School of Psychology, yWelsh Hearing Institute, University Hospital of Wales, Cardiff, Wales,

and zExecpt, Nancy, France

Accepted for publication 15 October 2001

Z H AO F. S T E P H E N S D. & M E Y E R-B I S C H C.

(2002) Clin. Otolaryngol. 27, 4–10

The Audioscan: a high frequency resolution audiometric technique and its clinicalapplications

The Audioscan is a form of high definition audiometry based on iso-hearing level frequency sweeps, which was

developed by Meyer-Bisch in 1990. Compared with traditional tone audiometry, it sweeps across the preset

frequency range at a predetermined sweep rate and provides a continuous audiometric curve. Because the device has

a maximum frequency range of 125–16 000 Hz with 64 frequencies per octave, the Audioscan method can,

theoretically, give 64 times as many values as fixed-frequency audiometry, which may give greater accuracy and

sensitivity. An advantage of this is its capacity to detect mild audiometric deficits such as narrow notches

situated between the frequencies normally tested. These may represent very limited auditory lesions, at a stage

when they cannot be detected by routine audiological methods. Thus, the Audioscan method can give not only a

detailed audiometric curve, but also provide important indicators of mild auditory dysfunction. The Audioscan

device (e.g. Essilor model) is commercially available for clinical purpose. It is a software-based system, which can

also be used for pure tone audiometry and Bekesy audiometry. This paper reviews the general aspects of the

Audioscan technique and current applications for detecting auditory dysfunction. This would be valuable to provide

some guidelines on the Audioscan assessment, and contribute to a clarification of the clinical application of

Audioscan and facilitate further research.

Keywords Audioscan notches noise-induced hearing impairment tinnitus genetic hearing impairment

King-Kopetzky syndrome

The Audioscan technique

ba s ic pr i nc i pl e s

The principle of the Audioscan technique is to provide a

frequency sweep at a constant hearing level across frequencies

instead of an intensity sweep through individual frequen-

cies.1,2 Thus, the sweep becomes horizontal instead of ver-

tical. The ranges of frequencies swept and levels explored are

determined by a programmed system which can be preset by

the tester. The starting level sweep rate and step size can also

be programmed. The logic of the algorithm is based on the

detection of audiometric irregularities, and defining them in

frequency and intensity. Therefore, the Audioscan provides

high sensitivity in the exploration of auditory deficits.

proc edur e

The first frequency sweep has a constant starting level. It starts

by just testing at octave intervals over the preset range until it

reaches a level at which there is a response. The subject

presses a button when he/she hears the stimulus and for as long

as he/she hears it. This information is recorded in the memory

of the machine. The programme then produces stimuli of the

next intensity and tests only in those parts of the frequency

range at which there was no response. The stimulus level is

subsequently further increased until the subject has responded

throughout the defined frequency range.

Clin. Otolaryngol. 2002, 27, 4–10

4 # 2002 Blackwell Science Ltd

Correspondence: Dr Fei Zhao, School of Psychology, CardiffUniversity, PO Box 901, Cardiff CF10 3GY, Wales (e-mail:[email protected]).

During Audioscan testing, the tone is presented at 1 kHz and

then goes on to high frequencies. Once at the maximum

frequency, the 1 kHz tone is retested before sweeping to low

frequencies. For example, say that the test intensity is initially

presented at 0 dB HL and the sweep frequency scan between

250 and 8 kHz revealed a range of frequencies between Fa1 and

Fb1 at which the signal was not heard. Thus tones between

250 Hz and Fa1 Hz, together with those between Fb1 and 8 kHz

were heard throughout the sweep by the subject. To complete

the audiogram, the device calculates the central frequency Fm¼(Fa1�Fb1)1/2. The sweep then starts again at Fm with a signal

level of 5 dB HL. The sweep is initially towards the higher

frequencies and then towards the lower (Fig. 1). This results in

the definition of two new values of Fa2 and Fb2. The process

then starts again at increasing steps of 5 dB until a continuous

response is obtained between Fa2 and Fb2. An Audioscan

audiogram may have a number of notches as shown in Fig. 2.

pa r am et e r s

For Audioscan testing the following parameters can be

adjusted by programming the device:

� Full frequency range: 125–16 000 Hz

� Starting level: between �10 dB and þ50 dB HL

� Sweep rate: between 7 and 99 s/octave

� Start side: right or left earphone

� Stimulus style: pulsed tone or continuous tone

� Step size: 1–10 dB

Therefore, a range of protocols can be chosen depending on

the specific measurement objectives sought.

Interpretation of the Audioscan results

As the Audioscan can test up to 64 frequencies within an

octave, it can provide detailed information on threshold

changes within that octave range. Moreover, the other impor-

tant information contained in an Audioscan is linked to the

presence of notches, which have been considered to be

indicators of mild auditory dysfunction.

notch m ea sur em ent

The parameters of notch measurement were recommended by

Laroche and Hetu,3 and are as follows (Fig. 3): the Centre

Frequency (Fa) corresponds to the frequency of the deepest

point of the notch; the Starting point (Sa): is the best hearing

threshold (dB HL) preceding the notch; The Absolute Value

in dB (Na) represents the absolute value of the notch lower

extremity (expressed in dB HL) measured at Fa; the Depth

(Pa) is the difference in dB between Na and Sa; and the Width

at 50% of depth (Wa) is determined using 50% of the total

notch depth; it is measured as a proportion of an octave.

Figure 1. The procedure of the sweeptechnique in the Audioscan test.

Figure 2. An example of an Audioscanaudiogram.

Techniques and applications of the Audioscan 5

# 2002 Blackwell Science Ltd, Clinical Otolaryngology, 27, 4–10

work i ng de f i n i t i on of audio scan notch

Using the Audioscan technique, a significant notch was

defined as being 15 dB or deeper than the surrounding fre-

quencies with the notch width not entering into the criterion.4

Factors influencing the results

The Audioscan technique is a type of semiautomatic audio-

metry. The ranges of frequencies swept and levels explored

are preprogrammed and then controlled by the patient’s

responses. Like all other psychoacoustical methods, a number

of factors affect the accuracy of the results of an Audioscan

test, for example, background noise, familiarity of the subject

with the method, concentration, alertness and duration of the

test, etc. In particular, the starting point of the response

depends very much on the test conditions and familiarity of

the subjects with the method. Therefore, to avoid the influence

of such factors, it is necessary to give the patients practice

listening to the Audioscan test before starting the test proper.

To overcome any initial unfamiliarity with the system on the

part of the patient, it has been suggested that the programme

should be stopped and then restarted after the subject

responds, discarding the initial response. If the subject under-

stands the normal test procedure and follows immediately by

releasing or pressing the button, the test procedure will then be

continued. Otherwise, if he/she does not release or press the

button, he/she will be carefully reinstructed, and then the test

procedure will begin again. This is important to improve the

reliability and accuracy of the test.

The reliability and sensitivity of Audioscan testing

The reliability and sensitivity of the Audioscan has been

evaluated systematically by Laroche and Hetu,3 who focused

mainly on the reliability and sensitivity of Audioscan thresh-

olds by comparison with the results obtained with fixed-

frequency Bekesy audiometry, when the sweep speed was

set at 10 or 20 s/octave, and the step size at 5 dB. The

measurement error (standard error) typically falls between

3.5 and 4.5 dB, a range of values which compares well with

those for Bekesy audiometry under optimal conditions. In

their study, high correlation coefficients between Bekesy and

Audioscan thresholds supported the validity of Audioscan

testing.

Furthermore, a pilot study by Meredith4 investigated the

reliability of the Audioscan notches as a function of the sweep

rate. The results indicated that Audioscan notches were found

less frequently in a group of three subjects when tested with

fast sweep rates (10 and 20 s/octave) than at slower sweep

rates (30 and 40 s/octave). In the study of Zhao et al.5 the

reliability of notches detected with a sweep rate of 30 s/octave

was significantly better than that at 15 s/octave. Nearly 50% (5

out of 12) of subjects failed to show notches at all when tested

at 15 s/octave although all had notches when tested at 30 s/

octave. This indicates that the reliability with the slower

sweep rate was better than that with the fast rate. This is in

keeping with the findings reported by Laroche and Hetu,3 who

found that the test-retest reliability was affected by sweep

speed, in that the reliability with their slower sweep rate (20 s/

octave) was better than that with fast sweep rate (10 s/octave).

The prevalence of Audioscan notchesin the control subjects

Various studies have investigated the prevalence of Audioscan

notches in normal control subjects. However, there is a

relatively wide range prevalence of Audioscan notches in

normal subjects among these studies, ranging from 13 to

25% in the frequency range up to 3000 Hz and 36–50% in

Figure 3. The parameters of notch mea-surement on the Audioscan.


6 F. Zhao et al.

the range up to 8000 Hz. These differences are mainly because

of a variety of influencing factors, for example, different

stimulus parameters and interpretative criteria used, test con-

ditions (e.g. background noise levels), and variability of the

experience of the subjects being tested. In addition, the

differences may also be influenced by the different genetic

backgrounds of the subjects tested.6

Demographic factors have been considered in the study by

Stephens et al.7 who argued that more data should be collected

in a large control group, which includes a normally distributed

age range of both genders. Subsequently a series of normative

results was collected from 70 normally hearing controls who

were selected on the basis of one per year age group per gender

from age 16 to 50.5 In this study, the prevalence of Audioscan

notches was 35.7% in the frequency band between 250 and

8000 Hz. Out of these, a total of 11 subjects (15.7%) had

notches between 500 and 3000 Hz, and 16 subjects had notches

between 3001 and 8000 Hz (22.9%) (two subjects with

notches in both frequency bands). These results are generally

in agreement with the findings in the previous studies.8–10

Furthermore, Fig. 4 shows the distributions of notches in the

two different frequency bands analysed by age band. In the

band from 3001 to 8000 Hz, there was a trend for the pre-

valence of notches to increase with the age band in each

gender. It demonstrates the accumulative influence of noise

exposure and other exogenous factors on the auditory system

of the individuals. However, in the 500–3000 Hz band, a stable

percentage of the notches was found in each age band and each

gender. This finding may imply that the notches in this region

represent a susceptible region in the cochlea influenced by

endogenous factors (e.g. a genetic factor).

There was no significant difference in the percentage of the

notches between males and females in the control group.

Moreover, no gender effects were found in the prevalence

of Audioscan notches either in the frequency band of 500–

3000 Hz or in the frequency band of 3001–8000 Hz. This

result is in accord with the finding by Stephens et al.,7 which

showed no significant sex effects on Audioscan notches in

normally hearing control subjects.

Clinical application of the Audioscan test

scr e en i ng f or no i s e - i nduced h ea r i ng

impa i rm ent

Meyer-Bisch1 used the Audioscan to detect small notches in

the hearing of those subjected to occupational noise. The

narrow notches found in the 3–8 kHz zone corresponded to

occupational noise-induced hearing impairment and other

disorders due to loud noise. He argued that it provided an

accurate indication of early hearing abnormalities in patients

exposed to noise. In the study of Laroche and Hetu,3 the

authors also confirmed the usefulness of Audioscan testing for

early identification of notches in the screening for noise-

induced hearing impairment. Screening for mild hearing

impairment owing to noise is a very important application

for the Audioscan method.

In an epidemiological survey on 1500 subjects who were

exposed to loud amplified music, such as personal cassette

players (PCP), discotheques and/or, rock concerts, a statisti-

cally significant deterioration of the average hearing thresh-

olds was found in young people using a PCP> 7 h/week

compared with those using one 2–7 h/week and compared

with their matched controls.11 The same trend was found in

subjects who went to rock concerts at least twice a month

compared with their matched controls. This suggests therefore

that the Audioscan method can provide sufficient accuracy

and sensitivity to measure very fine hearing deficits at an early

stage where prophylactic measures may be able to stop the

development of debilitating hearing impairment.

a ppl icat ion to t i nn i tu s pat i ent w ith

normal h ea r i ng

A total of 26 patients complaining of tinnitus and with normal

hearing on pure tone audiometry were investigated with the

Audioscan in the study of Sirimanna et al.; 96% (25 out of 26)

had notches between 250 and 8000 Hz.12 Moreover, matched

tinnitus pitch corresponded to the Audioscan notch within one

Figure 4. Percentage of subjects withAudioscan notches by age and gender, intwo frequency bands, in normally hearingsubjects.



octave in 24 out of the 26 subjects (92%). In all but one, the

Audioscan notches were found on the same side as the tinni-

tus ear. It may be postulated that such a tinnitus population

had a very high prevalence of localized damage to the inner

ear. Testing with transient-evoked otoacoustic emissions

(TEOAEs), showed that 75% (18 out of 24) subjects had

poor TEOAEs. This suggested that, even though the popula-

tion studied here had normal hearing thresholds, they had

subclinical cochlear (outer hair cell) damage, supporting the

hypothesis of peripheral tinnitus generation.

a ppl icat ion to carr i e r s of g en et ic

h ear i ng im pa i rm ent

Usher syndrome

Meredith4 first employed the Audioscan to detect carriers of

genes for hearing impairment to identify the audiometric

characteristics that might represent some carrier status of

certain types of genetic hearing impairment. He examined

different stimulus parameters and found a sweep rate of 30 s/

octave over the frequency range 500–3000 Hz and that a notch

size criterion of 15 dB or more gave optimal results in

differentiating the subjects from non-carriers.

When the test was administered to obligate carriers of

Usher syndrome type II, 100% of obligate carriers (five cases)

were found to have notches as were 43% of possible carriers

(14 cases), which is compatible with theoretical estimation of

50% risk of carrying the genes.13 Moreover, the number of

notches present in a similar normal population was only about

13.3%, which was similar to the estimation of Chung et al.14

of the carriers for non-syndromal recessive genetic hearing

impairment in the normal population. Therefore, the authors

concluded that the Audioscan technique might offer a sensi-

tive technique for the detection of carriers in families with

Usher syndrome type II.

However, other authors have obtained less consistent results

but have used different stimulus parameters and criteria for

notches.9,15 The study by Wagenaar et al.15 failed to replicate

the Usher II findings in patients with Usher syndrome type I.

Although in eight out of 50 control ears (16%) a notch was

found, the Audioscan method demonstrated notches in the

500 Hz�3 kHz range in only three out of the 10 (30%) of

obligate carriers. Differences between the results of this study

and compared with the findings of Meredith et al.13 might be

as a result of the fact that two different populations are being

compared. It seems possible that there is a real difference

between carriers of Usher type I and Usher type II syndromes.

So far, six different gene locations have been described for

Usher 1 syndrome and three for Usher 2,16 so that some of the

differences may be explained by the genetic heterogeneity. In

addition, there were several differences in the stimulus para-

meters and criteria used between the two studies (Table 1).

Table 1. Comparison of the findings of Audioscan notches in different studies on carriers of genetic hearing impairment

Genetic condition

Criteria of Notches Parameters Findings

DepthFrequencyband Width

Frequencyrange Sweep rate

Startinglevel

Obligatecarriers

Possiblecarriers

US type IIMeredith13 15 dB 0.5–3 kHz n.d. 300–4000 30 s/octave �5 dB 5 (5)¼ 100% 6 (14)¼ 43%Laoide-Kemp10 15 dB 0.5–6 kHz 1 octave 250–8000 30 s/octave n.s. 0 (4)¼ 0%�

2 (4)¼ 50%y3 (4)¼ 75%z

US type IWagenaar15 15 dB 0.5–3 kHz 250–8000 15 s/octave n.s. 3 (10)¼ 30%Laoide-Kemp10 15 dB 0.5–6 kHz 1 octave 250–8000 30 s/octave n.s. 3 (4)¼ 75%�

3 (4)¼ 75%y4 (4)¼ 100%z

NSARHLMeredith4 15 dB 0.5–3 kHz n.d. 300–4000 30 s/octave � 5 dB 25 (45)¼ 56%Stephens7 15 dB 0.5–3 kHz n.d. 300–4000 30 s/octave � 5 dB 7 (12)¼ 58% 10 (15)¼ 67%Cohen9 15 dB 0.25–8 kHz n.d. 250–8000 30 s/octave 5 dB below

the bestthreshold

6 (24)¼ 25%� 6 (8)¼ 75%

13 (24)¼ 54%§Laoide-Kemp10 15 dB 0.5–6 kHz 1 octave 250–8000 30 s/octave n.s. 11 (32)¼ 34%� 20 (49)¼ 41%�

16 (32)¼ 50%y 28 (49)¼ 57%y28 (32)¼ 75%z 31 (49)¼ 63%z

Lina-Granade6 15 dB 0.5–3 kHz n.d. 300–4000 30 s/octave n.s. 5 (14)¼ 36%Alport syndrome

Sirimanna17 15 dB 0.5–3 kHz n.d. 250–8000 30 s/octave �5 dB 8 (8)¼ 100% 8 (15)¼ 73%

Frequency bands: �0.25–3 kHz; y3–4 kHz; z4–6 Hz; §3–8 kHz.n.d.¼ not defined, n.s.¼ not specified.


8 F. Zhao et al.

This might also be the reason why these two studies have

come to different conclusions. A fast sweep rate (15 s/octave)

was used in the study by Wagenaar et al.,15 as mentioned

above, and this would be expected to lead to less reliable

results, with fewer notches being detected.

Alport syndrome

Table 1 shows the findings of subsequent studies in the

different clinical groups. Sirimanna et al.17 studied obligate

(female) carriers of Alport syndrome with no clinical hearing

impairment. They found that, out of eight tested, four had

hearing impairments and the other four had notches on the

Audioscan. Similarly, among 15 possible female carriers,

three had hearing impairments, five had notches on Audioscan

and seven had normal hearing. These results support the

findings studied by Meredith et al.13 in carriers of Usher

syndrome type II in terms of the sensitivity of the test to minor

changes in hearing levels.

Non-syndromal autosomal recessive

hearing impairment

In the study by Stephens et al.7 on non-syndromal autosomal

recessive hearing impairment, the authors reported that 58%

of obligate carriers had significant notches, as did 67% of

possible carriers. Such notches were more common in the

carriers of definite genetic hearing impairments than in normal

control subjects. This observation indicated that the Audio-

scan could be a useful tool used in the detection of carriers of

autosomal recessive genes for hearing impairment.

However, studies by Cohen et al.9 and Laoide-Kemp et al.10

indicated that the Audioscan notches were not significantly

more common in the carriers of non-syndromal autosomal

recessive hearing impairment than in control subjects (20%

versus 25%, and 25% versus 34%, respectively). There were,

however, several differences in the stimulus parameters com-

pared with those in the study by Stephens et al.7 although they

adopted similar criteria for the notches. Variable results in the

audiological detection of carriers of non-syndromal autosomal

recessive genetic hearing impairment could also be explained

by the complex heterogeneity of the condition, with 30

different genes having been located at the time of writing.17

Furthermore, the action of many of the genes responsible for

non-syndromal autosomal recessive hearing impairment

(NSARHI) is in parts of the cochlea (e.g. stria vascularis),

which is unlikely to result in the localized effects necessary to

result in a narrow notch. Genes acting directly on the hair cell

are most likely to have such an effect. A precise protocol and

criteria for the notches is crucial for comparing the findings of

different studies, and such a protocol and criteria have recently

been described.18

In addition, future studies on this heterogenous population

group should be restricted to those individuals in whom the

family genotype is known.

audio scan t e st i ng i n pat i ent s w ith

k i ng^kopetzky sy ndrom e (ob scur e

auditory dys fu nct ion)

King–Kopetzky syndrome is defined as the condition in which

an individual complains of difficulties understanding speech

in the presence of background noise but has normal hearing

thresholds on pure tone audiometry.19,20

Compared with the prevalence of Audioscan notches in the

control group, significantly higher percentages of notches

were found in patients with King–Kopetzky syndrome in

the 500–3000 Hz range. However, there was no significant

difference in the percentage of notches in the 3001–8000 Hz

frequency band between the King–Kopetzky syndrome and

control groups. This indicates that the main peripheral impair-

ment basis of King–Kopetzky syndrome may be concentrated

on the region of 500–3000 Hz. Moreover, there was a sig-

nificantly higher percentage of notches in this region when

analysed by age band and gender in the King–Kopetzky

syndrome than in the control group. These results further

indicate that Audioscan notches may represent a fine hearing

deficit as an indicator of mild cochlear abnormality in patients

with King–Kopetzky syndrome.

A significantly higher percentage of subjects with Audio-

scan notches between 500 and 3000 Hz was found to have a

family history of hearing impairment than those without

notches. However, there were no significant differences in

the percentages of family history of hearing problems between

subjects with and without Audioscan notches at 3001–

8000 Hz in the two groups. These results confirm that Audio-

scan notches between 500 and 3000 Hz in patients with King–

Kopetzky syndrome may be associated with genetic hearing

disorders. In addition, such notches were significantly wider

than those found in carriers of Usher II syndrome, NSARHI

and in control subjects, and we have argued that this is

suggestive of a subclinical representation of a dominant

genetic disorder,21,22 preceding the mid-frequency and wider

notches often associated with such conditions.

Conclusions

This review has discussed current knowledge regarding the

Audioscan technique and its clinical application. With Audio-

scan audiometry, early signs of hearing abnormalities can be

detected in terms of threshold changes within the octave range

and/or narrow notches. Well defined parameters of notches

will be useful for the investigation and comparison of the

characteristics of the notches in different clinical contexts.

Furthermore, Audioscan notches in the different frequency

zones might have some diagnostic relevance. Various studies

have confirmed that the commonest findings are the detec-

tion of such narrow notches in the 3001 to 8000 Hz zone

in individuals exposed to noise or ototoxic substances.



Moreover, Audioscan notches in the frequency range of 500–

3000 Hz are commonly associated with genetic hearing dis-

orders, and the Audioscan may be capable of identifying mild

auditory dysfunction, which may be influenced by the genetic

background. The mechanism is unclear but it may be the more

susceptible zone to genetic disorders or degenerative pro-

cesses affecting the cochlea.

The main weakness of the Audioscan is that unless large

step sizes are used, the test duration in patients with significant

hearing losses can be long, thus reducing its reliability. We

would, thus, recommend use of the Audioscan in patients with

clinically normal hearing, but the use of sweep frequency

Bekesy audiometry in those in whom a hearing loss is

expected. This can be achieved simply by a change in the

menu of the Audioscan programme.

References

1 MEYER-BISCH C. (1990) Audiometrie automatique de depistagepreventif: le balayage frequentiel asservi (Audioscan). CahiersNotes Documentaires 139, 335–345

2 MEYER-BISCH C. (1996a) Audioscan: a high-definition audio-metry technique based on constant-level frequency sweeps: anew method with new hearing indicators. Audiology 35, 63–72

3 LAROCHE C. & HETU R. (1997) A study of the reliability ofautomatic audiometry by the frequency scanning method(Audioscan). Audiology 36, 1–18

4 MEREDITH R. (1991) Audiometric identification of carriers ofnon-manifesting genes for deafness. MSc Thesis, University ofSouthampton

5 ZHAO F., STEPHENS S.D.G., MEREDITH R. et al. (1998) Parameteranalyses of Audioscan notches in carriers of genetic hearing loss.In Developments in Genetic Hearing Impairment I, pp. 60–67.Stephens, D., Read, A. & Martini, A. (eds). Whurr, London

6 LINA-GRANADE G., KREISS M., GELAS T. et al. (1998) Cochlearirregularities in obligate carriers of recessive genetic hearingimpairment and in controls. In Developments in Genetic HearingImpairment I, pp. 68–73. Stephens, D., Read, A. & Martini, A.(eds). Whurr, London

7 STEPHENS D., MEREDITH R., SIRIMANNA T. et al. (1995)Application of the Audioscan in the detection of carriers ofgenetic hearing loss. Audiology 34, 91–97

8 SIRIMANNA T. (1993) Identification of carriers of Alport’ssyndrome by audiometry. MSc Thesis, University of Manchester,Manchester, UK

9 COHEN M., FRANCIS M., LUXON L.M. et al.. (1996) Dips onBekesy or Audioscan fail to identify carriers of autosomalrecessive non-syndromic hearing loss. Acta Otolaryngol. 116,521–527

10 LAOIDE-KEMP S., PARKER D., PRATT S. et al. (1996) Audio-metric identification of carriers of non-syndromal autosomalrecessive genes for hearing loss. Paper presented at the 23rdInternational Congress on Audiology, Bari, Italy

11 MEYER-BISCH C. (1996b) Epidemiological evaluation of hearingdamage related to strongly amplified music (personal cassetteplayers, discotheques, rock concerts): high-definition audio-metric survey on 1364 subjects. Audiology 35, 121–142

12 SIRIMANNA T., STEPHENS D. & BOARD T., (1996) Tinnitus andAudioscan notches. J. Audiol. Med. 5, 38–48

13 MEREDITH R., STEPHENS D. & SIRIMANNA T. (1992) Audio-metric detection of carriers of Usher’s syndrome type II. J.Audiol. Med. 1, 11–19

14 CHUNG C.S., ROBINSON O.W. & MORTON N.E. (1958) A note ondeaf mutism. Ann. Hum Genet 23, 357–366

15 WAGENAAR M., SNIK A.F.M., KIMBERLING W.J. et al. (1996)Carriers of the Usher syndrome type I: is audiometricalidentification possible? Am. J. Otolaryngol. 17, 853–858

16 VAN CAMP G. & SMITH R.J.H. (2001) Hereditary Hearing LossHomepage. World Wide Web URL: http://dnalab-http://www.uia.ac.be/dnalab/hhh

17 SIRIMANNA K.S., FRANCE E. & STEPHENS S.D.G. (1995)Alport’s syndrome: can the carriers be identified by audiometry?Clin. Otolaryngol. 20, 158–163.

18 STEPHENS D., LINA-GRANDE G., HATZOPOULOS S. et al. (1999)Audiometric investigation of carriers. In Developments inGenetic Hearing Impairment I, pp. 33–35. Stephens, D., Read,A. & Martini, A. (eds). Whurr, London

19 SAUNDERS G.H. & HAGGARD M.P. (1989) The clinicalassessment of obscure auditory dysfunction. Ear Hearing 10,200–208

20 STEPHENS D. & RENDELL R.J. (1988) Auditory disability withnormal hearing. Quaderni Audiologia 4, 233–238

21 ZHAO F. & STEPHENS D. (1999) Audioscan testing in patientswith King–Kopetzky syndrome. Acta Otolaryngol. 119,306–310

22 STEPHENS D. & ZHAO F. (2000) The role of a familyhistory in King–Kopetzky Syndrome. Acta Otolaryngol. 120,197–200


10 F. Zhao et al.

Documents

The Audioscan: a high frequency resolution audiometric technique and its clinical applications