TITLE: BASELINE OTOACOUSTIC (BOA) TEST

Hearing tests basically fall into two categories: 1) behavioural testing with subject describing or indicating what they hear and 2) biophysical tests of hearing variables where the information is measured with sensors; mostly these are electric potential measurements or acoustic measurements using microphones. However, in theory this information could be obtained by patient imaging or use of magnetic field detection equipment. The second class of measurement may be of naturally occurring phenomena from which hearing parameters are determined, or they may be (mostly are) determined by evoking a response, i.e. perturbing the ear in some way and measuring a response to that stimulus. Evoking stimuli for the hearing organ mostly are sounds, but they may also be by electrical stimulation (e.g. in the cochlear prosthesis known as the Bionic ear), or it may be through mechanical vibration, or motions which may evoke vestibular (dizziness) response. Sound-evoked ^' biophysical ear responses are now commonly obtained by playing sounds or clicks to the ear and measuring "echoes" from the ear with a microphone sealed in the ear canal. These, are widely known as evoked otoacoustic emissions (EOAE), literally sounds emitted from the ear, and these are believed to originate from 'muscle-like' activity of the outer hair cells (OHC) in healthy cochleae. The name given to this phenomenon is the "cochlear amplifier" because it appears that these OHC preamplify the sound signals ("ac" vibration) for easier detection. Deafness results from loss of this internal amplification. Like electronic amplifiers which need a "dc" source of voltage, so also does the mammalian ear. (This internal battery drops its voltage as we age and this is thought to underlie age-related hearing loss). That these EOAE are not seen in deaf ears, indicating loss of amplification, forms the basis of current clinical tests, and the measuring apparatus was originally marketed under the original patents held by Dr. David Kemp of the Institute of Laryngology and Otology in London which have all lapsed.

EOAE are soft sounds but they provide a very great deal of information about the state of the ear emitting them. EOAE may be not just recorded for how much response they would give for any standard stimulus entering the ear, they can also be recorded to record how much these sounds can be suppressed (or more generally modulated) by other stimuli. In general when EOAE are discussed, by far the majority of users would equate this as being the same thing as making a 1) transient-evoked response (TEOAE), 2) a distortion product response (DPOAE) or a 3) stimulus frequency response (SFOAE). A sound goes into the ear and a sound comes out of the ear and the analysis is performed on the sound that comes out. In general, EOAE are regarded as being small signals buried in noise and so advanced signal processing techniques are required to extract the signal from the noise. The common methods of processing are signal averaging and frequency filtering. For DPOAE the filters may be set very narrow, to analyse the microphone signal at only the frequency where it is expected. This is not applicable for TEOAE where the transients are typically Click stimuli (CEOAE) and generate relatively wide-band responses. For this method most of the noise is considered to be at low frequencies and so the signal is typically high-pass filtered at ~700 Hz, removing all the low frequency components. Extensive signal averaging is used to extract the remaining frequency components. In general, EOAE are considered to contain frequencies which are audio frequencies and till recently, no allowance has been made for very low frequencies (<700 Hz) which may not be entirely contaminating noise. The latest research publications are describing a low-frequency distortion product (the quadratic distortion product, QDP) which does occur at low frequencies, but no allowance is yet made

that this component may belong to a larger class of low-frequency signals better described as shifts in the signal baseline which represents the cochlea adapting to stimulus change. The very earliest reports of baseline shifts in cochlear mechanical phenomena stem from LePage in 1987. While the structure of the cochlea necessarily depends upon a biological battery and 'dc' potentials which in turn influence the activity of the OHC and their 'dc' (mechanical) operating point the field of investigators have not yet embraced the notion that not only audio-frequency sounds are emitted, but there are signals measureable in the ear canal which are so low frequency they do not fall into the class of emissions within the audio frequency spectrum. The fact that a microphone can detect a baseline pressure change in the ear canal may indeed not be a "sound" being emitted at all, but evidence of a mechanical impedance change within the cochlea which the microphone registers as a low-frequency signal along with the higher frequency emission components.

What is new about this patent: BASELINE OTO ACOUSTIC TEST (BOA)

This patent concerns making a measurement on the ear which may not necessarily entail a "sound" emitted from the ear. The subtle but important distinction here made is that most people think about sounds as being within the audible frequency range i.e. 20 Hz to 20 kHz. If a pressure variation occurs outside of this frequency band, the average person would not necessarily equate that with a sound because it is virtually inaudible. Sounds may be inaudible because they are ultrasonic (frequencies above 20 kHz) or infrasonic (frequencies below 20 Hz). The sounds in question maybe infrasonic, or near infrasonic, i.e. very low frequencies less than 100 Hz. These sounds are not necessarily thought of as "being emitted" from the ear. In making this test the standard probe in the ear canal may be primarily registering changes which are occurring in the inner ear and the results may be a far clearer description of the kinds of information being sought by conducting an otoacoustic emission test.

While the communication between the cochlea and the acoustic probe in the ear canal is via the middle ear, the ear drum and resulting air pressure waves, these waves are so low in frequency that they are delivering information about how the static pressure or pressure baseline is changing. It is thought that important processes controlling the sound coding process in the inner ear are indeed occurring at very low frequencies. These processes are called homeostatic processes, or simply homeostasis. The mechanical measure of homeostasis is the operating point (OP) of the OHC or, in conventional terminology, its "dc" or present value of mechanical displacement. Since the OHC are continually adapting to changes in signal strength this value changes, so that for example, increasing the stimulus strength may cause the operating point to move unidirectionally positive, and vice versa. This mechanical change is reflected in the ear canal as either a low-frequency sound or more generally the result of an intracochlear pressure change. Such direct (intracochlear) measurements have been made for very many years, but only now are investigators seriously considering that not only "ac" sound signals are measureable in the ear canal, but "dc" signals as well. This being the case, just as the "ac" emissions have been useful diagnostically, the "dc" signals may be also. Indeed if the "dc" signals or measures of OP change indeed reflect cochlear control signals then they potentially provide a far more potent clinical indicator of the health of the ear.

Basic hearing scientists know well that the OHCs are not just amplifiers, they are also regulators. They slowly (and continually) perform mechanical regulation of hearing sensitivity as one moves from a quiet situation to a noisy situation and back again. The

sophistication of this regulation process is remarkable, BUT TO DATE nobody has thought that these infrasonic movements within the cochlea could be directly assessed with an otoacoustic emission probe. It has always been assumed that in order to measure intracochlear changes you must obtain the internal information via the standard otoacoustic EMISSION approaches. We now know that regular EOAE analysed only for their power and phase may provide ambiguous information depending on the value or movement of the operating point. Being able to obtain a direct measure of the operating point and its polarity changes gives much less ambiguous information about signals controlling cochlear sensitivity.

This patent application concerns the making of a measurement of homeostatic processes in the inner ear WITHOUT being limited to a specific frequency associated with any member of the family of distortion products. While for 25 years the cubic distortion product (CDP) has been the one deemed the distortion product which gives the most useful information about cochlear status, it is being realised that the QDP may give better information. This patent generalises the low-frequency requirement still further by not constraining the baseline variations to a specific frequency (and perhaps some very restricted model for its origin), but rather observing the whole baseline variation as it occurs in the ear canal and giving a new range of clinical diagnoses directly in terms of OHC adaptation and/or descending control signals from the brain acting on the OHC.

We are not measuring sounds which are emitted from the ear in the conventional sense. We are merely measuring baseline pressure variations which we believe are more instantaneously coupled to the microphone sensor. The technical innovation in this patent concerns overcoming the severe technical problems in making such a measurement. These baseline pressure variations are extremely small and dominated by other very low frequency sounds, such as breathing noises, and the pressure pulse due to contractions of the heart.

These measurements are made under computer control and use specific tricks in signal processing to extract the baseline pressure variation. It comes down to asking the relevant question, and in turn that is gained from a detailed knowledge of cochlear physiology (not possessed by many hearing professionals in clinical practice). At the point of submitting this application one of the many possibilities has been investigated and it can be said that the direct homeostatic approach yields a far simpler description of relevant processes than does EOAE, viz. the ^' low-frequency signals previously deemed noise-components contaminating the EOAE are not noise at all, but measures of OHC operating point which has a totally different time history.

In essence not even top hearing scientists yet appear to have taken this approach. The provisional patent application (30/3/2005) was submitted before a key experiment was presented to the scientific community. More explicit details were revealed at the ARO meeting in February 2006 and drew comments of much interest. The presentation describes how the regulation of the operating points of the OHC vary with interplay of two tones played to the ear. It is well known that this information will give important diagnostic information, previously only obtainable in animal experiments, or obtainable indirectly via evoked otoacoustic emissions with multiple possible interpretations.

POINT SUMMARY OF DESCRIPTION

1. The prior art of otoacoustic emission methodology filters out important information by processing the signal from the microphone in the ear canal with a sharply tuned filter set to a particular distortion product frequency or by high-pass filtering the whole response to remove noise and using extensive signal averaging. The new algorithm takes account of the time course of the baseline pressure in the ear canal which is a better indication of the underlying status of cochlear activity, of adaptive response and of homeostatic pressures which set the ear's sensitivity and vulnerability to trauma and which is otherwise used to diagnose hydrops, condition underlying Meniere's attacks. Accordingly the claim is that it is possible to more directly assess control processes in the cochlea, rather than the prior art of obtaining indirect measures, in how the measurement is carried out upon the ear and in how the response of the ear is processed. The claim therefore is for new hardware and measurement algorithms specifically engineered to assess the operational status of the ear.

2. As referred to in Summary 1, the prior art followed the traditional approach which believed that because we hear only sounds in the frequency range 20 Hz to 20 kHz, that otoacoustic signals being emitted from the human ear only contain those frequencies also. The prior apparatus used to collect these signals only made allowance for this, by filtering out signals outside of this range, indeed by filtering out signals thought to be noise obscuring the emission signal. The new methodology recognises that not all of these "noise" components of the recorded signal are noise unrelated to cochlear activity, but are instead large signals with a time course not allowed for in previous recording algorithms. The new methodology analyses these very low frequency components by recognising that to regard them as having a specific frequency such as a particular distortion product, is again to conform to the prior art. More generally the apparatus analyses baseline variation without necessarily expecting that this variation has a particular frequency.

3. The accompanying drawing illustrates that the BASELINE OTOACOUSTIC (BO A) frequency band is below the regular evoked otoacoustic emission frequency band; it is so low in frequency that it may be considered NOT an emission at all in regular usage and therefore all claims are not bound by any existing patents for Otoacoustic Emissions (per se).

4. As detailed in preceding points, the otoacoustic signal being recorded from the probe in the ear canal contains both signals related to the cochlear activity and homestatic . regulation and signals which are genuinely unrelated such as cardiac pressure pulses. Prior art regarded all these signals as "noise" which must be removed by filtering or by use of extensive signal averaging to improve the signal to noise ratio (SNR). The claim is that from understanding the underlying nature of the "noise" it is possible to totally revolutionise the data collection algorithm. The prior art gave signals and responses which are clearly strongly regulated, but which yielded only indirect clues as to how they were being regulated.

5. As detailed in point 4, the new methodology allows the separation of genuine "noise" from baseline variation of the signal. This was treated as noise previously, but now is regarded as vital information regarding the history of variation of the operating point of the active process centred upon the outer hair cells.

6. As detailed in point 5, the genuine noise which is primarily heart beat affecting the baseline pressure is removed by playing a two-tone burst signal to the ear using a masking algorithm, then shortly later, replaying this digitally-generated stimulus with its polarity reversed. This allows computing the differential response which largely removes cardiac pressure pulses and middle ear muscle contractions from the baseline. The remaining signal

is then employed as a direct measure of outer hair cell control processes and baseline variation previously treated as noise and removed by filtering.

7. In respect of point 6, middle ear muscle contractions pull on the eardrum and affect the baseline pressure. Due to their characteristics these contractions are readily recognised and the new measurement algorithm taken their activity into account by two approaches a) removing those responses during middle ear muscle activity from the data being collected, b) by keeping the stimulus sound levels lower than the higher levels normally required to evoke these contractions and c) by relating the incidence and nature of these contractions to the resulting EOAE.

8. As summarised above, the Baseline OtoAcoustic (BOA) algorithm directly estimates the OHC mechanical operating point directly via the baseline air pressure in the ear canal, rather than the prior art, which uses distortion product (either cubic or quadratic) components of the signal to describe that measure of cochlear status indirectly. The BOA approach is much more general and includes important frequency components not contained in the distortion product family of discrete frequencies.

9. Since the BOA algorithm provides a more direct assessment of OHC condition, modulation of that condition by neural pathways from the brain to the ear, may be more easily assessed, and this may be of great advantage when assessing hearing in children with complex hearing problems which potentially include "auditory neuropathies" and "central auditory processing disorders" which, because of the descending pathways cannot be regarded as exclusively central, but may occur because of a subtle cochlear involvement.

10. This methodology is useful to diagnose Meniere's disease via indicators of intracochlear hydrops (with its operating point change) by virtue of directly registering an operating point change.

11. In respect of the OP shift mentioned above, the methdology also enhances estimations of risks for (susceptibility to) hearing loss for those exposed to loud noise, adding this approach to the set of strategies being used to reduce noise-induced hearing loss. The prior art is to register a change in emission strength at a particular frequency or range of frequencies which is not place-specific. However, as a function of time may be used to estimate individual human susceptibility to hearing loss based upon the operating point shift due to a challenge tone.

12. Further to point 11, since it is possible to assess the effect of any noise exposure due to temporary shift of the baseline, this new methdology has application to hearing loss prevention to assess how stable are the operating points of the any individual ear (or pair). D