Susan J. Norton, Ph.D. - US grants

The goal of the proposed research is to obtain a systematic set of data on the input-output relations among evoking stimuli and evoked oto-acoustic emissions in large numbers of normal and non-normal human ears. The proposed research will provide a systematic description of evoked oto-acoustic emissions in normal and non-normal ears and provide a basis for evaluating their clinical applicability. Evoked emissions appear to be highly tuned and subject to both forward and simultaneous masking and suppression. These characteristics of evoked emissions, together with their susceptability to metabolic and traumatic insults, and their non-linear relation to stimulus amplitude suggest that they originate within the cochlea and are highly related to both the frequency selective and non-linear processes of the cochlea. The exact site and/or mechanism of their generation, as well as their true distribution across ears and thei significance, are matters of some controversy. At least some of these controversies are due to the lack of systematic data. Despite the large numbers of papers that have appeared in the literature since Kemp's (1978) original work, relatively few conditions and subjects are actually reported. Systematic delineation of the relations between emissions evoked by different stimuli and the relations between the status of the peripheral auditory system and characteristics of evoked emissions is important if their source and significance are to be identified. The proposed research will provide a systematic description of evoked oto-acoustic emissions in normal and non-normal ears and provide a basis for evaluating their clinical applicability.

The goal of the proposed research is to determine the potential clinical utility and significance of evoked oto-acoustic emissions. Evoked emissions appear to be closely linked to non-linear, frequency-selective elements within the cochlea. In addition, they are highly sensitive to cochlear status and can be altered easily by such metabolic and traumatic insults as intense sound exposure, ototoxic drugs and anoxia. These characteristics of evoked emissions suggest that they are a potential non-invasive tool for assessing cochlear status in humans. The proposed research will provide a systematic set of data on the input-output relations among evoking stimuli and evoked oto-acoustic emissions from large numbers of normal and non-normal human ears. These data will provide a basis for evaluating the clinical applicability of emissions. The proposed research will also provide insight into the source and/or mechanisms involved in the generation of emissions, as well as insight into alterations in cochlear status such as permanent sensorineural hearing loss, Meniere's disease, ototoxic drugs and intense noise exposure. A unique feature of the proposed research is the application of signal analysis techniques that have been used in other fields (e.g. sonar, radar, radiology, geophysics) where recovery of low-level signals embedded in noise is also a goal to the measurement and analysis of evoked oto-acoustic implemented in such a way as to be clinically accessible.

The primary goal of the proposed research is understanding otoacoustic emissions and their relation to cochlear mechanics in normal and abnormal human ears. Otoacoustic emissions provide a direct, non-invasive measure of outer hair cell function, cochlear nonlinearities and the role of auditory efferents in cochlear function. Thus, they are a potentially powerful non-invasive research and clinical tool for assessing normal and abnormal cochlear mechanics in humans. However, several questions concerning the nature of the emission generator(s) and its relationship to cochlear status must be addressed in order to optimally interpret clinical data. Specific questions to be addressed by the proposed research include: (1) What is the nature of the emission generator ? (2) What are the relationships between transient-evoked, stimulus-frequency and combination-tone emissions ? (3) What is the role of auditory efferents in normal cochlear mechanics ? (4) How are emissions, and thus indirectly cochlear mechanics, affected by temporary threshold shift and permanent hearing loss due to noise exposure and aging in the absence of other cochlear insults ? The proposed research will provide a systematic set of data on the input-output relations among evoking stimuli and evoked emissions in large numbers of normal and abnormal ears. These data will provide a basis for better understanding the nature of cochlear hearing loss and the clinical potential of otoacoustic emissions, and serve as data for evaluating and formulating models of normal and abnormal cochlear mechanics.

The importance of early intervention to minimize the handicapping effects of hearing impairment makes early diagnosis desirable. Unfortunately, the widespread screening of neonates has been hampered by the lack of sufficiently robust, reliable and cost-effective tools, and by the relatively low prevalence of hearing impairment in this population. The central goal of this multi-center collaborative application is to obtain a large unbiased database containing information about the stimulus-response characteristics of three physiologic responses related to hearing status - transient evoked otoacoustic emissions (TOAE), acoustic distortion product emissions (ADP) and auditory brainstem responses (ABR) - in normal-hearing and hearing-impaired neonates. Four thousand eight hundred (4800) NICU infants and 2400 normal well-baby nursery infants will be evaluated at six sites using ABR, TOAE and ADP. Neonatal hearing status will be validated by behavioral responses to auditory stimuli at 8-12 months corrected age. The accuracy of each tool for identifying neonatal hearing impairment, as validated by visual reinforcement audiometry, will be determined using methods based on signal detection theory. The primary measure of accuracy will be the area under the receiver operating characteristic (ROC) curve. ROC analysis provides a description of disease detectability that is independent from both disuse prevalence and decision threshold effects. These data will be used to design screening protocols with known accuracy and costs as a function of target population and setting. Prototype model systems will be developed and evaluated. Ultimately, these data in concert with modem technology, can be used to design inexpensive, intelligent devices to implement universal screening for hearing impairment. The influence of age and medical status on the performance of TOAE, ADP and ABR in identifying peripheral hearing impairment, as well as the cost of implementation and cost-effectiveness will also be evaluated.