Search

EP-3429457-B1 - METHOD FOR AUTOMATICALLY DETERMINING AN INDIVIDUAL FUNCTION OF A DPOAE LEVEL MAP OF HUMAN OR ANIMAL HEARING

EP3429457B1EP 3429457 B1EP3429457 B1EP 3429457B1EP-3429457-B1

Inventors

  • DALHOFF, ERNST
  • ZELLE, Dennis

Dates

Publication Date
20260513
Application Date
20170314

Claims (15)

  1. Method for the automated determination of an individual function of a DPOAE level map with p dp,I = f ( L 1 , L 2 ) of human or animal hearing, wherein distortion product otoacoustic emissions (DPOAEs) are presented as acoustic pressure p in dependence on levels L 1 and L 2 of the two primary tones used for the generation of DPOAEs, characterized in that it comprises the following steps: - reading (110) of a model function (70) p dp,M = f ( L 1 , L 2 ) with model parameters of a DPOAE level map, based upon a number of N DPOAE measurements of a stimulation frequency pair f 1 , f 2 with respectively different level pairs { L 1 (1... N ) , L 2 (1... N ) } in a population (p) of normally hearing subjects into a main memory (15) of a computer unit (10), wherein N ≥ 40 and p ≥ 2, - automatically presenting (120) n different level pairs { L 1 (1... n ) , L 2 (1... n ) } of a stimulation frequency pair f 1 , f 2 via tone output means (21, 22) to an individual and detecting the corresponding DPOAEs of the individual via tone recording means (23), wherein at least the first level pair { L 1 (1) , L 2 (1) } is predefined, and wherein n << N, - iteratively adapting (130) the model function p dp,M = f ( L 1 , L 2 ) to the measured n DPOAEs until an individual function p dp,I = f ( L 1 , L 2 ) is obtained with individual parameters of a DPOAE level map of the individual by the computer unit (10), and - outputting (140) the individual function p dp,I = f ( L 1 , L 2 ) and/or its individual parameters to an output means (11) of the computer unit (10), wherein the method is characterized in that the model function has a more or less linearly rising ridge (73) to which more or less linearly linked level pairs { L 1 (G) , L 2 (G) } are assigned, wherein at least half of the measured level pairs { L 1 , L 2 } lie at least 5 dB remote from either side of the group of the level pairs { L 1 (G) , L 2 (G) } assigned to the ridge (73).
  2. Method according to claim 1, characterized in that the first level pair { L 1 (1) , L 2 (1) } has a level L 1 (1) of 67 ± 10 dB and a level L 2 (1) of 57 ± 10 dB.
  3. Method according to any one of the above claims, characterized in that the different level pairs { L 1 , L 2 } are presented in a sequence that is identical for each individual.
  4. Method according to any one of the above claims, characterized in that the predefined, different level pairs { L 1 , L 2 } are presented in a sequence that has a number of k subsequences whose level pairs { L 1 , L 2 } are essentially transverse to the linearly linked level pairs { L 1 (G) , L 2 ( G ) } assigned to the ridge.
  5. Method according to any one of the above claims, characterized in that n is ≥ 5 and ≤ 12, and preferably n is ≥ 5 and ≤ 8.
  6. Method according to any one of the above claims, characterized in that k ≥ 2 and ≤ 8.
  7. Method according to any one of the above claims, characterized in that the level pairs { f 1 (2... nk ) , L 2 (2.. nk ) } of a subsequence following a first predefined level pair { L 1 (1) , L 2 (1) } is determined with n k measurements by a function { L 1 ( i ) , L 2 ( i ) } = { L 1 ( i -1) + µ · Δ L 1 , L 2 ( i -1) + µ · Δ L 2 } from the respectively preceding level pair { L 1 ( i- 1) , L 2 ( i -1) }, wherein µ = ±1, preferably +1, and Δ L 1 , Δ L 2 is a level distance of two sequential level pairs and has values of Δ L 1 = 4 to 14 dB, preferably from 6 to 10 dB - and Δ L 2 = 0 to -2.78 dB , preferably Δ L 2 = -1.52 to -2.78 dB.
  8. Method according to any one of the above claims, characterized in that , when the first level pair { L 1 (1) , L 2 (1) } and the second level pair { L 1 (2) , L 2 (2) } produce two DPOAEs with p dp,I (12) that each have a signal-to-noise ratio of >= 4 dB, preferably >= 10 dB, the level of a subsequent third level pair { L 1 (3) , L 2 (3) } is adjusted to differ by at least Δ L 1 ≥4 dB from the level of the preceding level pair { L 1 (2) , L 2 (2) } when p ap,I (2) - p dp,I (1) > 0; on the other hand, the level of a subsequent level pair { L 1 (3) , L 2 (3) } is adjusted to differ by at least Δ L 1 ≤ -4 dB from the level of the first level pair { L 1 (1) , L 2 (1) }, when p dp,I (2) - p dp,I (1) ≤ 0.
  9. Method according to any one of the above claims, characterized in that , when the first level pair { L 1 (1) , L 2 (1) } does not produce any DPOAEs with p dp,I (1) that have a signal-to-noise ratio of ≥ 4 dB, preferably ≥ 10 dB , the same search direction is continued until either the maximum or minimum stimulation level L 1 ( i ) is reached, or a group of three valid DPOAEs with p dp,I ( i..i +2) was produced that have each a signal-to-noise ratio of ≥ 4 dB , preferably ≥ 10 dB.
  10. Method according to any one of the above claims, characterized in that , when a group of three valid DPOAEs that have a signal-to-noise ratio of ≥4 dB, and preferably ≥10 dB is not produced in the first subsequence, another subsequence is started with a higher level pair { L 1 ( i +1) , L 2 ( i +1) }, wherein the start level pair for the new subsequence is set to L 2 ( i +3) = L 2 (1) + 20 ± 10 dB, L 1 ( i +3) = L 1 (1) + 20 ± 10 dB, or at most to the maximum achievable level.
  11. Method according to any one of the above claims, characterized in that , after detecting the DPOAEs of at least 3 level pairs { L 1 (1..3) , L 2 (1..3) }, preferably of one subsequence, the position of the ridge { L 1 (G) , L 2 ( G ) } along the line formed by the three level pairs is determined from these three level pairs { L 1 (1..3) , L 2 (1..3) }, and a fourth level pair { L 1 (4) , L 2 (4) } is presented that is placed at a predetermined distance down the ridge, wherein the group average of the ridge direction ϕ is used, and wherein a slope of the linear ridge of the level map is calculated using the DPOAEs determined from the four presented level pairs { L 1 (1..4) , L 2 (1..4) ).
  12. Method according to claim 1 or 11, characterized in that , when a group of three valid DPOAEs with p dp,I i- 2 ..i is produced in the first or second subsequence that each have a signal-to-noise ratio of >=4 dB, preferably >=10 dB, the level pair below the ridge { L 1 (G) , L 2 ( G ) } = L 1 ( i -2) + ε · Δ L 1 , L 2 ( i -2) + ε · Δ L 2 } is determined by adapting a mathematical function to the associated DPOAEs p dp,I i -2... i , wherein ε must be calculated so that p dp,I ( L 1 ( G ) , L 2 ( G ) ) forms a maximum, and on that basis a fourth level pair L 1 ( i +1) , L 2 ( i +1) is presented, with a function { L 1 ( i +1) , L 2 ( i +1) } = { L 1 ( i ) + Δ L 1 , L 2 ( i ) + Δ L 2 }, wherein Δ L 2 = -15 ± 10 dB is adjusted, and the level pair is preferably adjusted to the projection of the anticipated ridge on the L 1 , L 2 level, i.e., adjusted with ΔL 1 /ΔL 2 ≈ 0.51 ± 0.15, and wherein a slope m of the approximately linear ridge of the level map is determined using the DPOAE calculated from the four presented level pairs L 1 ( i -2 ...i +1 ), L 2 ( i -2... i +1) .
  13. Method according to any one of the above claims, characterized in that the level pairs { L 1 (1.. n ), L 2 (1.. n ) } are presented as pulsed, wherein each individual pulse is presented with a duration T D of 2 to 40 ms.
  14. Method according to claim 13, characterized in that the level pairs { L 1 (1. .n ) , L 2 (1.. n ) } are presented within a measuring block consisting of a plurality of level pairs { L 1 (1.. n,m ) , L 2 (1.. n , m ) } which are presented sequentially over time in pulses, wherein level pairs { L 1 (1.. n,i ) , L 2 (1 ..n,i ) }; { L 1 (1.. n,i +1) , L 2 (1.. n,i +1) } that follow each other directly in time have different stimulation frequencies { f 2 ,i , f 1, i }; { f 2 ,i+ 1 , f 1, i +1 }, wherein, in particular, in an additional method step, the determined individual function of a DPOAE level map and its parameters are stored by the computer unit (10) in a non-volatile memory (16).
  15. A system configured for performing the method according to any one of the above claims 1 to 14, with a computer unit (10), a main memory (15), a non-volatile memory (16) for storing a model function p dp, M = f ( L 1 , L 2 ) and model parameters of the model function, at least one tone output means (21, 22) controlled by the computer unit (10) for presenting tones to an individual, with at least one tone recording means (23) connected to the computer unit (10) for detecting DPOAEs from the ear of the individual, wherein in particular the at least one tone output means (21, 22) is a speaker with a highly linear characteristic and/or wherein in particular an output means (11) is provided for outputting the individual function of a DPOAE level map to a user.

Description

The present invention relates to a method for automatically determining an individual function of a DPOAE level map of a human or animal ear according to claim 1, and to a system for carrying out this method according to the dependent claim. The present invention relates in particular to a method comprising individual features of claim 1, and to a system for carrying out this method comprising individual features of the dependent claim. The auditory system can be viewed as a chain of successive signal processing blocks. These blocks are traversed before the more complex perception of hearing arises in the cortex. The first blocks of the signal processing chain are the outer ear (auricle and ear canal), the middle ear (ossicles with the footplate forming the boundary to the fluids of the inner ear), and the fluid-filled inner ear. The vast majority of hearing loss originates in the inner ear. This includes age-related hearing loss, which on average leads to a 25 dB hearing loss in women and 35 dB in men between the ages of 60 and 70 in the frequency range above 4 kHz. It is primarily caused by an impairment of the so-called cochlear amplifier, which, in a healthy state, amplifies incoming sound waves by a factor of 300–1000 before they are converted into neural signals by the inner hair cells and their synapses. Since around 1980, the existence of the cochlear amplifier has been gradually proven, with David T. Kemp's discovery of otoacoustic emissions (OAEs) playing a central role in this. These are sounds that are generated as a byproduct by the active amplifier and transmitted backward through the middle ear to the ear canal. There, they can be measured with sensitive miniature microphones. One form of OAE is distortion product otoacoustic emissions (DPOAE), in which two primary tones with frequencies f1andf2 and levels L1 and L2 are typically presented. The nonlinear characteristic of the mechanoelectric transduction of the ion channels of the outer hair cells, which constitute the main motor element of the cochlear amplifier in humans and mammals in general, leads to numerous distortion products. The most easily measurable distortion product, and therefore the one preferred in diagnostic applications , is that at f dp = 2 f1 - f2 with f2 > f1 and an optimal frequency ratio of approximately f2f1=1,2 Nowadays, DPOAEs are usually stimulated such that, at the cochlear imaging site of the second primary tone, both excitation frequencies result in vibration amplitudes of the basilar membrane that are as equal as possible. Accordingly, diagnostic conclusions from DPOAE findings are interpreted at the frequency and excitation level of the second primary tone { f2 , L2 }. DPOAE measurements can be performed , for example, according to the method described in the DE 102014108663 The described procedures are performed and interpreted at different frequencies. A well-known method for determining the hearing threshold, or more precisely the threshold of the cochlear amplifier of the inner ear, is based on determining the threshold above which an emission, in particular a distortion product otoacoustic emission (DPOAE), is measurable, a technique that was first used in humans [ BP Kimberley and DA Nelson., J. Otolaryngol., 18(7): 365-369, 12 1989 ; DA Nelson and BP Kimberley, J. Speech Hear. Res., 35(5):1142-1159, 10 1992 ]. For this purpose , a growth function is typically measured, e.g., the emissions with excitation levels of L₂ = 60 dB SPL decreasing in 5 dB steps until a certain SNR (signal-to-noise ratio) is no longer achieved. The excitation level at which the required SNR is just barely reached is then referred to as the DPOAE threshold. According to Boege and Janssen [ P. Boege and T. Janssen, J. Acoust. Soc. Am., 111(4): 1810-1818, 04, 2002 ] the values of the growth function are plotted semilogarithmically, that is, the DPOAE are plotted linearly as sound pressure in units of [µPa] on the ordinate, against the excitation sound pressure of the second primary tone L 2 in the logarithmic unit of [dB SPL]. With so-called optimal excitation sound pressures [ P. Kummer, T. Janssen, P. Hulin, and W. Arnold. Optimal L1-L2 primary tone level separation remains independent of test frequency in humans. Hear.Res., 146(1-2):47-56, 08 2000 Typically, this yields a linear growth function, which can be extrapolated to the abscissa using linear regression. In this case, the extrapolated intersection point of the growth function and the abscissa is called the estimated DPOAE threshold (also: "estimated distortion product threshold", EDPT). In general, this DPOAE threshold correlates well with the psychoacoustically measured thresholds [ P. Boege and T. Janssen, J. Acoust. Soc. Am., 111(4): 1810-1818, 04, 2002 , MP Gorga et. al., J. Acoust. Soc. Am., 113(6): 3275-3284, 06, 2003 .], but exhibits an unsatisfactory high standard deviation and in some cases a deviation of up to 40 dB [ N. Schmuziger et. al., J. Acoust. Soc. Am.,