Dept. of Biomedical Engineering
Oregon Health & Science University
This dissertation research is focused on developing new noninvasive methods of inferring physiologic processes using features of otoacoustic emissions (OAEs). The physiologic processes that were studied include blood glucose levels, focused auditory attention, and age-related hearing changes. Current state-of-the art methods for inferring such physiologic processes are invasive and oftentimes painful, time-consuming and labor-intensive, or non-existent. OAEs are low intensity sounds generated by the cochlea in response to acoustic stimuli. Evoking and measuring an OAE is done using a tiny speaker and microphone that fit snugly inside the ear canal. The OAE response can be partially inhibited or reduced in amplitude by presenting competing acoustic stimuli contralaterally (opposite ear), ipsilaterally (same ear), or both. This inhibition effect is caused by activation of the medial olivocochlear (MOC) efferent nerve pathways from the brain. In this dissertation, results are presented that show how stimulus frequency (SF) OAE amplitudes and phase as well as MOC-related changes in SF OAEs correlate with blood sugar levels. A mechanism is proposed whereby blood glucose acts (1) as a metabolic mediator of the endocochlear potential and (2) as a neurotransmitter by being transformed to ATP and acting on ATP-sensitive ion channels in the cochlear outer hair cells; a process which is mediated by calcium release by activation of the MOC. A new prediction methodology is described that is called Hyperglycemic Risk Analysis (HRA) in which elevated blood sugar levels are predicted using metrics from SFOAEs recorded with and without activation of the MOC. Leave-one-out cross-validated receiver operating characteristic (ROC) analysis showed HRA achieves good accuracy with an area-under-the-curve (AUC) of 0.85. Clinical applications for HRA could ultimately lead to a method for noninvasively monitoring elevated glucose levels in diabetes patients. Also included within this dissertation are results from an experiment in which we investigated how focused attention influences MOC inhibition. Results are presented from an experiment in which test subjects simultaneously had their MOC activated while they performed psychoacoustic listening tasks that required varied levels of focused attention. Features within the MOC inhibition were analyzed relative to the psychometric functions to infer effect of focused attention on the MOC. A statistically significant increase in the MOC inhibition amplitude was observed when the subjects were actively participating in the listening task compared with when the subjects were passively listening. Finally, this dissertation presents results demonstrating how aging influences the auditory system and specifically SF OAE amplitude and latency. Results from a 53-subject study are presented that show how SF OAE amplitudes can be used to predict age-related hearing changes to the auditory system. A new relationship between SF OAE probe frequency and suppressor level are also presented and a corresponding model is described. Results are presented as they relate to proposed extensions to a current auditory model.
School of Medicine
Jacobs, Peter Gordon, "Inference of physiologic processes using otoacoustic emissions" (2010). Scholar Archive. 646.