John J. Rosowski

MEASUREMENTS OF MIDDLE EAR PRESSURE GAIN AND COCHLEAR INPUT IMPEDANCE IN THE CHINCHILLA

Measurements of middle ear conducted sound pressure in the cochlear vestibule PV have been performed in only a few individuals from a few mammalian species. Simultaneous measurements of sound-induced stapes velocity VS are even more rare. We report simultaneous measurements of VS and PV in chinchillas. The VS measurements were performed using single-beam laser-Doppler vibrometry; PV was measured with fiberoptic pressure sensors like those described by Olson [JASA 1998; 103: 3445-63]. Accurate in-vivo measurements of PV are limited by anatomical access to the vestibule, the relative sizes of the sensor and vestibule, and damage to the cochlea when inserting the measurement device. The small size (170 μm diameter) of the fiber-optic pressure sensors helps overcome these three constraints. PV and VS were measured in six animals, and the middle ear pressure gain (ratio of PV to the sound pressure in the ear canal) and the cochlear input impedance (ratio of PV to the product of VS and area of the footplate) computed. Our measurements of middle ear pressure gain are similar to published data in the chinchilla at stimulus frequencies of 500 Hz to 3 kHz, but are different at other frequencies. Our measurements of cochlear input impedance differ somewhat from previous estimates in the chinchilla and show a resistive input impedance up to at least 10 kHz. To our knowledge, these are the first direct measurements of this impedance in the chinchilla. The acoustic power entering the cochlea was computed based on our measurements of input impedance. This quantity was a good predictor for the audiogram at frequencies below 1 kHz. Thesis Supervisor: John J. Rosowski Title: Professor of Otology & Laryngology and Health Sciences & Technology, Harvard Medical School and Harvard-MIT Division of Health Sciences & Technology

Pressures in the human cochlea during bone conduction

The mechanisms of bone conduction (BC) hearing, which is important in diagnosis and treatment of hearing loss, are poorly understood, thus limiting use of BC. Recently, information gained by intracochlear pressure measurements has revealed that the mechanisms of sound transmission that drive pressure differences across the cochlear partition are different for air conduction (AC) than for round-window stimulation. Presently we are utilizing these pressure measurement techniques in fresh human cadaveric preparations to improve our understanding of sound transmission during BC. We have modified our technique of intracochlear pressure measurements for the special requirements of studying BC, as bone vibration poses challenges for making these measurements. Fiberoptic pressure sensors were inserted through cochleostomies in both scalae at the base of the cochlea. The cochleostomies were then tightly sealed with the sensors in place to prevent air and fluid leaks, and the sensors were firmly secured to ensure u...

Dynamics of the tympanic membrane: Multiple-specimen study of digital holograms

Stroboscopic digital holography has been used to measure sound-induced tympanic membrane (TM) surface motion with a high spatial resolution. In the current state of the art, holograms from different specimens can be compared qualitatively by inspection and quantitatively by manual identification of regions of interest. However, anatomical variations in the shape of the TM and geometrical variations due to changes in relative position and orientation of the specimen with respect to the camera preclude point-by-point metrics across specimens. The aim of this study is to create a set of shape-normalized TM motion maps in order to quantify the average motion and variability in a set of specimens. We present a method in which the motion maps of 5 cadaveric human TMs were rotated, translated, scaled and sheared to normalize TM orientation, size, and position, and we show preliminary results of cross-specimen analysis of motion.

Design of a mechatronic positioner for a holographic otoscope system

We are developing an advanced computer-controlled digital optoelectronic holographic system (DOEHS) with the ability to measure both shape and acoustically induced deformations of the tympanic membrane of several species, including humans. The DOEHS have been deployed for testing and use in clinical environment. The DOEHS consists of laser delivery (LD), optical head (OH), sound presentation (SP), computing platform (CP), and a mechatronic otoscope positioner (MOP) subsystems. In this paper, we present advances in our development of the MOP subsystem, which is capable of positioning the OH subsystem near the patient’s ear and maintaining of its relative position and orientation during holographic eardrum examinations. Our work focuses on the development and implementation of various approaches for mechanical stabilization of the MOP-OH subsystems, including custom designed packaging of the OH as well as automatic interferometric compensation against measuring disturbances induced by periodic oscillations, such as those produced by heartbeat and breathing of a patient during examination. We present preliminary results of our investigations of acoustically induced motions on tympanic membranes by measurements with our DOEHS enabled with a structurally stable MOP subsystem.

Adaptative reconstruction distance in a lensless Digital Holographic Otoscope

We are developing a Digital Optoelectronic Holographic System (DOEHS) for measurements of shape and deformations acoustically induced of the human tympanic membrane (TM) in the clinic. Such measurements will be used to perform quantitative diagnosis of the middle-ear. The DOEHS platform consists of laser-delivery illumination (IS), optical head (OH), image-processing computer (IP), and positioning arm (PS) subsystems. Particularly, the OH of the DOEHS has been configured as an in-line, lensless, holographic arrangement to quantify deformations of the TM when it is subjected to controlled sound excitation. Holographic information is recorded by a digital camera and reconstructed numerically by the Fresnel integral approximation. Accurate measurements are achieved when TM samples are imaged within the depth of focus of the OH, which requires the selection of the optimal numerical reconstruction distance. In this paper, we discuss an experimental approximation to identify the best reconstruction distance based on the position of the reference beam (RB) within the OH and its associated phase factor within the Fresnel integral approximation. Results of these investigations are being used to further optimize the OH design of the DOEHS system, and to improve the numerical reconstruction algorithms used. Representative experimental measurements are shown.