ThesisPDF Available

Morphological modeling and motion measurements of the middle ear using new X-ray stereoscopic and tomographic techniques

Authors:

Abstract and Figures

The measurement of three-dimensional motion is important to study the mechanics of biological systems. For decades, efforts have been made to quantify and measure them using different approaches. X-ray stereoscopy as a technique was introduced one year after Roentgen discovered x-rays. The method is used to obtain 3D coordinates of marker points, and derive their displacement. Continuous changes in position and velocity of a marker along its path of motion can also be studied, but to my knowledge this has not been done before. Thus I studied the 3D motion of internal non-visually exposed structures using x-rays, which is not possible with other available techniques. I performed x-ray stereoscopy using a single x-ray point source, unlike the classical setups with two sources. This method was tested on a well-controlled test object to measure static displacements in the range of 0.5 mm. The obtained amplitude of motion was verified using laser Doppler vibrometry. The results showed a displacement uncertainty of less than 5 μm. Although this method shows promising results in measuring motion amplitudes, it is still not usable to quantify the velocity of a marker point. Therefore, this method was combined with a greyscale analysis to study the motion in an accurate way: A moving object with a relative high x-ray absorbance leaves a trace in an x-ray shadow image of which the greyscale intensities are linearly proportional to the marker velocity at that point. If the marker moves fast then less x-rays are absorbed and vice versa. The integrated intensity along the marker linear path of motion can be used to reconstruct the time information. To do so, I placed tungsten beads, with high x-ray absorption, on a calibration object to reconstruct the path of motion as a function of the known input waveform. Measurements on a test object showed that the path of motion can be reconstructed with an accuracy of about 5% compared to a sinusoidal input waveform. My method shows that the 3D motion can be studied using a single x-ray point source within a few seconds. A big difference compared to a CT-scan, which requires a 360 degree scan of the object and intensive calculations afterwards. Moreover, it can also be used for measurements on non-visually exposed internal structures, which is not possible with other methods such as moiré interferometry or laser Doppler vibrometry. The developed method is used to study the 3D motion of the middle ear ossicles as a function of high quasi-static pressure variations. The latter occur on a daily basis, and it is known that they have an influence on hearing thresholds. The effects of these high pressure variations on the ossicles are not entirely clear due to the complex anatomical structure of the middle ear. The 3D motion of the ossicles of gerbil (malleus) and rabbit (malleus and incus) were studied as a function of peak-to-peak pressures of 1 up to 2 kPa at frequencies of 0.5 Hz to 50 Hz. The displacement of the ossicles in both gerbil and rabbit as a function of pressure has a sigmoid (S-like) shape. The displacement increases faster at lower pressures, while it increases much slower and stagnates when higher pressures are applied. Thus, approximately 80% of the displacement occurs in the range of 0 to 1 kPa range, and only a limited increase of the amplitude is observed at higher pressures (from 1 to 2 kPa). The ossicles also have a smaller peak-to-peak amplitudes in the static and quasi-static range, compared to frequencies closer to the acoustic domain. The ossicles motions are visualized on high-resolution computer models to show the movement in these pressure regimes. Ears are subject to different sound pressures, with high or low intensity and changing rapidly or slow. The tympanic membrane plays a key role in regulating these pressure variations. To get a better understanding of the middle ear operating conditions, I developed a setup to study the tympanic membrane regulatory capacity in the quasi-static pressure regime. The pressure at the middle ear side was measured as a function of ear canal pressurization in the high quasi-static pressure regime. The trans-tympanic pressure difference was found to be the smallest in the lower quasi-static range, and quickly increasing as a function of frequency. The non-linearity of the middle ear pressure as a function of ear canal pressure was investigated and found to increase as a function of frequency. The tympanic membrane can protect the middle ear for more than 60% of high pressure amplitudes, such as a 2 kPa peak-to-peak, to which the outer ear is exposed. Finally, I created a library of high-resolution morphological computer models of the middle (and inner) ear of cat, gerbil, rabbit, rat and human (adult and juvenile) using state-of-the-art micro-CT. These models are of great importance for finite element modelling, which requires accurate geometries to obtain realistic results. The models are freely available at the laboratory of the biomedical physics group website for the benefit of the research community, clinicians and educators. The gerbil and rabbit models were furthermore used to visualise the 3D motion of the middle ear ossicles, the main aim of this dissertation.
Content may be subject to copyright.
Faculteit Wetenschappen
Departement Fysica
Morphological modelling
and motion measurements of the middle ear
using new X-ray stereoscopic and
tomographic techniques
Morfologische modellering
en bewegingsmetingen op het midden oor
met nieuwe X-stralen stereoscopische en
tomografische technieken
Proefschrift voorgelegd tot het behalen van de graad van
doctor in de Wetenschappen
aan de Universiteit Antwerpen
te verdedigen door
Wasil Hashim M. Salih
Promotor - Prof. Dr. Joris Dirckx
Co-Promotor - Dr. Jan Buytaert
Antwerpen, 2013
Contact information
B Wasil Salih
Laboratory of Biomedical Physics - Department of Physics - University
of Antwerp
Groenenborgerlaan 171 - 2020 Antwerp - Belgium
@ wasil.hashim@ua.ac.be
wasilhashim@yahoo.com
m http://www.ua.ac.be/bimef
H +32 (0) 488 83 04 61 (Belgium)
+249 (0) 9 129 79 156 (the Sudan)
ISBN 978-90-5728-405-2
Legal deposit D/2012/12.293/44
Copyright
c
by Wasil Salih 2013. All rights reserved. No part of this
dissertation may be reproduced or transmitted in any form or by any
means, electronic, mechanical, photocopying, recording, or otherwise,
without prior written permission of the author.
Cover illustration
A snapshot of a computer model of the gerbil middle ear. See page 36
for more details.
Doctoral committee
Supervisor
Prof. Dr. Joris J. J. Dirckx
Head of the Laboratory of Biomedical Physics (BIMEF), Department
of Physics, Faculty of Science, University of Antwerp, Belgium.
Co-supervisor
Dr. Jan A. N. Buytaert
Laboratory of Biomedical Physics (BIMEF), Department of Physics,
Faculty of Science, University of Antwerp, Belgium.
Chairman
Prof. Dr. Jan Sijbers
Vision Lab, Department of Physics, Faculty of Science, University of
Antwerp, Belgium.
Internal member
Prof. Dr. Sara Bals
Electron Microscopy for Materials Science (EMAT) Research Group,
Department of Physics, Faculty of Science, University of Antwerp, Bel-
gium.
External members
Prof. Dr. Magnus von Unge (MD)
Professor and Consultant, Department of Otorhinolaryngology,
Akershus University Hospital and University of Oslo, Norway,
Associated Professor, Department of Clinical Science, Intervention and
Technology, Karolinska Institute, Stockholm, Sweden.
Dr. Ir. Manuel Dierick
Department of Physics and Astronomy, Ghent University.
Private PhD Defense
Monday, December 10, 2012 at 10:00h
in room U338, Campus Groenenborger,
University of Antwerp.
Public PhD Defense
Friday, January 18, 2013 at 15:30h
in room U024, Campus Groenenborger,
University of Antwerp.
This PhD work, as with any piece of work that needs a sponsor, was not
possible without the financial support of the Sudanese Ministry of Higher
Education and Scientific Research, and Alneelian University, the Sudan.
The PhD candidate was partially financially supported by the Research Foun-
dation Flanders (FWO).
This work is dedicated to:
my parents, sisters and brother,
my wife and kids,
my relatives and friends,
and to everyone who suffers from a hearing impairment
for motivating me to perform this research.
Preface
Hearing research is of great importance to understand the causes of hearing
impairments, which affect millions of people worldwide. Sound is a pres-
sure wave in an elastic medium such as air, and can occur at very different
frequencies and amplitudes. The effects of sound on the ear are not straight-
forward to understand. Although much effort has been made to understand
the biomechanics of the ear, its complexity ensures that many questions re-
main.
The mammalian ear is divided into three anatomical parts: the outer, mid-
dle and inner ear. The combination of these parts allows us, for instance, to
enjoy music, make conversation with friends or avoid life endangering circum-
stances. In the most simple way, the function of the ear can be described as
follows: the outer ear collects sound waves and passes them to the eardrum,
which sets the small middle ear bones (ossicles) in motion; these transfer the
vibration to the fluid filled inner ear. There, the sound is transformed into
electrical pulses and transferred through the auditory nerve to the brain for
interpretation. This doctoral dissertation will focus on the middle ear.
The first intention of the research for this dissertation was to develop a
method to study three-dimensional motion within non-transparent objects:
the x-ray stereoscopy technique. This method was applied to study the 3D
motion of the ossicles under quasi-static pressure variations. Such pressure
variations occur throughout daily life, and are known to have an influence
on hearing thresholds.
To demonstrate the motion of the ossicles, high-resolution morphological
computer models are needed. Therefore, state-of-the-art middle (and inner)
ear models of different species were created from micro-computed tomog-
raphy scans. These models are also of great importance to finite-element
vii
Preface
modelling, a method very suitable to simulate the mechanical behaviour of
more complex systems such as the middle ear. The pressure in the middle
ear as a function of the outer ear pressure was also studied in this work; a
study that contributed to our knowledge of the tympanic membrane’s role in
pressure regulation.
This dissertation consists of six chapters. The first chapter gives a gen-
eral overview of the anatomy and physiology of the ear. It also includes a
discussion on the relevance of hearing research and the objectives and the
motivation of this doctoral research.
The second chapter presents high-resolution morphological computer models
of the middle (and inner) ear for different species, including humans. The
third chapter discusses the ability of the eardrum to regulate the pressure
at the middle ear side when the outer ear is exposed to high quasi-static
pressure variations.
In chapter four, x-ray stereoscopy is used to study 3D sub-millimetre dis-
placements of a static test object. In chapter five, a technique is presented
to study the 3D motion of non-visually exposed internal structures of an
object. The method combines classical x-ray stereoscopy with time informa-
tion obtained from greyscale variation in x-ray shadow images, resulting in
a new approach to study periodical motions inside non-transparent objects.
This technique is used to study the motion of the ossicles in the quasi-static
pressure regime in both gerbil and rabbit ears. The results of this study are
discussed in chapter six.
viii
Table of contents
Preface vii
1 General introduction 1
1.1 Ear anatomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1.1 Outer ear . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1.2 Middle ear . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.1.3 Inner ear . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.2 Ear function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
1.3 Sound pressure regimes . . . . . . . . . . . . . . . . . . . . . . . . 15
1.4 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
1.4.1 Hearing loss . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
1.4.2 Hearing research impact . . . . . . . . . . . . . . . . . . . . 23
2 High resolution 3D models 27
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
2.2 Materials and methods . . . . . . . . . . . . . . . . . . . . . . . . . 29
2.2.1 Dissection . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
2.2.2 X-ray imaging . . . . . . . . . . . . . . . . . . . . . . . . . 30
2.2.3 3D segmentation and reconstruction . . . . . . . . . . . . . 31
2.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
2.3.1 High-resolution 3D models . . . . . . . . . . . . . . . . . . 33
2.3.2 Morphological parameters . . . . . . . . . . . . . . . . . . . 38
2.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
3 Middle ear pressurization 47
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
3.2 Material and methods . . . . . . . . . . . . . . . . . . . . . . . . . 49
3.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
3.4 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
3.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
ix
Table of contents
4 X-ray stereoscopy 59
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
4.2 Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
4.3 Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
4.4 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
4.5 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
4.6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
5 Motion measurements with an x-ray technique 73
5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
5.2 Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
5.2.1 Measurement of 3D coordinates using cone beam stereoscopy 75
5.2.2 Measurement of the motion . . . . . . . . . . . . . . . . . . 75
5.3 Experiments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
5.3.1 Test object . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
5.3.2 Gerbil ear . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
5.4 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
5.4.1 Test object . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
5.4.2 Gerbil ear . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
5.4.3 Background correction . . . . . . . . . . . . . . . . . . . . . 86
5.5 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
5.6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
6 Ossicles motion 95
6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
6.2 Materials and method . . . . . . . . . . . . . . . . . . . . . . . . . 98
6.2.1 Sample preparation . . . . . . . . . . . . . . . . . . . . . . 98
6.2.2 Pressure generation . . . . . . . . . . . . . . . . . . . . . . 100
6.2.3 Measurement of the motion . . . . . . . . . . . . . . . . . . 100
6.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
6.3.1 Gerbil ear . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
6.3.2 Rabbit ear . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
6.3.3 3D displacement . . . . . . . . . . . . . . . . . . . . . . . . 106
6.4 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
6.4.1 Measurement setup . . . . . . . . . . . . . . . . . . . . . . . 115
6.4.2 Ossicles displacement . . . . . . . . . . . . . . . . . . . . . 116
6.4.3 3D motion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
6.4.4 Transfer function . . . . . . . . . . . . . . . . . . . . . . . . 119
6.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
General conclusion 123
Samenvatting 127
x
Table of contents
Arabic abstract 132
A Matlab codes 137
Curriculum vitae 155
List of publications 159
List of abbreviations 161
Acknowledgement 163
xi
List of Figures
1.1 The human ear . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 The human pinna . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3 The middle ear . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.4 Tympanic membrane outlines . . . . . . . . . . . . . . . . . . . . . 6
1.5 Human tympanic membrane . . . . . . . . . . . . . . . . . . . . . . 7
1.6 The ossicles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.7 Human inner ear . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.8 Human cochlea . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.9 Middle ear amplification . . . . . . . . . . . . . . . . . . . . . . . . 14
1.10 Sound pressure spectrum . . . . . . . . . . . . . . . . . . . . . . . 16
1.11 Mammals audible range . . . . . . . . . . . . . . . . . . . . . . . . 18
1.12 Hearing aids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
1.13 Middle ear implants . . . . . . . . . . . . . . . . . . . . . . . . . . 23
1.14 An example of a cochlear implant . . . . . . . . . . . . . . . . . . . 24
1.15 Hearing research input/output . . . . . . . . . . . . . . . . . . . . 26
2.1 Segmentation using Amira
R
package . . . . . . . . . . . . . . . . 32
2.2 Reconstructed cross-section image . . . . . . . . . . . . . . . . . . 34
2.3 Models of the ossicular chain for the different species . . . . . . . . 35
2.4 Model of a Gerbil middle and inner ear . . . . . . . . . . . . . . . 36
2.5 Malleus models for the different species . . . . . . . . . . . . . . . 37
2.6 Model of a gerbil malleus blood vessels . . . . . . . . . . . . . . . . 38
2.7 Measurement of the ossicles dimensions . . . . . . . . . . . . . . . 39
2.8 The behavioural upper frequency limit of hearing . . . . . . . . . . 42
3.1 Trans-tympanic pressure setup . . . . . . . . . . . . . . . . . . . . 50
3.2 Example of both the middle and inner pressure signal . . . . . . . 52
3.3 An ear pressurization with 500 Pa . . . . . . . . . . . . . . . . . . 52
3.4 An ear pressurization with 1000 Pa . . . . . . . . . . . . . . . . . . 53
3.5 An ear pressurization with 2000 Pa . . . . . . . . . . . . . . . . . . 53
xiii
List of Figures
3.6 Trans-tympanic pressure measurement . . . . . . . . . . . . . . . . 54
3.7 Normalized trans-tympanic pressure measurement . . . . . . . . . 55
3.8 The Total Harmonic Distortion of ME pressure signal . . . . . . . 56
4.1 Simple stereoscopy system . . . . . . . . . . . . . . . . . . . . . . . 63
4.2 Simple x-ray stereoscopy setup . . . . . . . . . . . . . . . . . . . . 67
4.3 Two stereo projections of a test object . . . . . . . . . . . . . . . . 69
5.1 X-ray stereoscopy setup to measure the motion of a test object . . 78
5.2 A function generator voltage signal measured with LDV . . . . . . 80
5.3 Gaussian distribution at the centre of a bead . . . . . . . . . . . . 82
5.4 X-ray shadow image of three Tungsten beads on top of a test object 82
5.5 Greyscale as a function of position . . . . . . . . . . . . . . . . . . 84
5.6 Path of a test object motion . . . . . . . . . . . . . . . . . . . . . . 85
5.7 X-ray shadow image of two beads at a gerbil malleus ME . . . . . 86
5.8 The motion of the Tungsten bead at the malleus of a gerbil . . . . 87
5.9 X-ray shadow image of one bead on a gerbil malleus with non-
constant background . . . . . . . . . . . . . . . . . . . . . . . . . . 87
5.10 A gerbil malleus displacement with corrected background . . . . . 88
5.11 Projected sphere diameter . . . . . . . . . . . . . . . . . . . . . . . 91
6.1 Picture of beads on a rabbit ossicles . . . . . . . . . . . . . . . . . 99
6.2 Custom-built pressure Generator . . . . . . . . . . . . . . . . . . . 101
6.3 Stereoscopy setup used for ears . . . . . . . . . . . . . . . . . . . . 102
6.4 Experimental setup at UGCT . . . . . . . . . . . . . . . . . . . . . 103
6.5 Shadow images for three beads on a gerbil malleus . . . . . . . . . 104
6.6 Displacement of gerbil umbos . . . . . . . . . . . . . . . . . . . . . 105
6.7 Average amplitude of gerbil umbos . . . . . . . . . . . . . . . . . . 106
6.8 Displacement of gerbil manubria . . . . . . . . . . . . . . . . . . . 107
6.9 Displacement of rabbit umbos . . . . . . . . . . . . . . . . . . . . . 108
6.10 Displacement of rabbit manubria . . . . . . . . . . . . . . . . . . . 109
6.11 Displacement of rabbit Stapes . . . . . . . . . . . . . . . . . . . . . 109
6.12 Average amplitude of rabbit ossicles . . . . . . . . . . . . . . . . . 110
6.13 Displacement of a rabbit (R4) malleus and stapes . . . . . . . . . . 111
6.14 Snapshot of the 3D motion in a gerbil . . . . . . . . . . . . . . . . 113
6.15 Snapshot of the 3D motion in a rabbit . . . . . . . . . . . . . . . . 114
xiv
List of Tables
1.1 Examples of sound pressure levels . . . . . . . . . . . . . . . . . . . 17
2.1 Parameters for the different species models . . . . . . . . . . . . . 40
2.2 Gerbil tympanic membrane parameters . . . . . . . . . . . . . . . . 43
2.3 Rabbit tympanic membrane parameters . . . . . . . . . . . . . . . 44
4.1 3D coordinates at rest position of a test object . . . . . . . . . . . 70
4.2 3D coordinates at 50 µm displacement of a test object . . . . . . . 70
4.3 3D coordinates at 100 µm displacement of a test object . . . . . . 70
5.1 The piston-like motion and the rocking-like motion measured with
LDV . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
xv
Chapter 1
General introduction
One of the best hearing
aids a man can have is an
attentive wife.
Groucho Marx, 2/10/1890 - 19/8/1977.
Throughout daily lives, hearing is involved in many of our activities such
as watching TV, listening to the radio, enjoying music from an mp3 player,
making a call to a friend, having a small talk with our family/friends during
the mealtime, or teaching our knowledge to our children. Moreover, knowl-
edge, traditions and customs were transmitted at early ages from generation
to generation verbally since writing and reading were not available to the
public. For all of these activities our ears serve to let us make contact with
our surroundings.
How does the ear work? This question has been posed since ancient times,
long before humans understood the concept of scientific research. The answer
is not as simple as most of the public thinks, but it can be generally answered
as follows: sound is collected by the outer ear (OE), it then travels through
the ear canal (EC) and causes the tympanic membrane (TM), or eardrum,
to vibrate. Then through the middle ear (ME) ossicles (the ME ossicles are
malleus, incus and stapes, cf. 1.1.2), these vibrations are transferred to the
inner ear (IE), where the vibrations are transformed into electrical pulses
and sent to the brain via the auditory nerve for interpretation, cf. figure 1.1.
1
1 General introduction
In the following sections, more detail about the morphology and physiology
of outer, middle and inner ear is presented.
1.1 Ear anatomy
The ear is divided, from the anatomical point of view, into three major parts
as shown in figure 1.1.
Figure 1.1: Schematic overview of the human ear morphology
(edited from Purves et al. (2003)).
1.1.1 Outer ear
The OE is the first part of the ear, it mainly consists of two parts. The first
is the auricle or pinna and the second is the auditory meatus or the EC, cf.
figure 1.1. The pinna, which is the external and the most lateral
1
part of
1
In anatomy, there are some terminologies to describe the directions, which are:
superior-inferior up and down
anterior-posterior front and back
medial-lateral inwards and outwards
2
1.1 Ear anatomy
the ear, has several prominences and depressions and consists of a thin plate
of cartilage covered with thin skin. The shape of the pinna varies between
different species, having been adapted to the environment. For instance,
the African elephant pinna has been adapted to act as a radiator-convector
to maintain its body temperature in the warm weather experienced in the
African plains (Buss and Estes, 1971). The pinna collects and amplifies
sound and passes it to the EC, which is a tube connecting the OE to the
ME. The EC consists of cartilage in the first third of its length; the rest is
a bony structure that is embedded in the temporal bone (TB), a dense hard
petrous structure. The EC is covered by a skin layer. With the help of a
waxy yellowish substance (earwax), the EC serves to protect the ear against
bacteria.
Figure 1.2: The anatomy of the human pinna (from Hunter and
Yotsuyanagi (2005)).
As shown in figure 1.2, the human pinna has several anatomical features. The
outer rim of the ear is called the helix. It extends from the most superior part
of the pinna and ends at the earlobe, which occupies the most inferior part
of the pinna. The earlobe has a soft connective tissue structure and varies
in size and shape between humans. The antero-inferior continuation of the
helix until the area just above the EC is called the crus helix. The antihelix
is a Y-shaped curved cartilaginous ridge, parallel to the helix. At its superior
part, the antihelix is split into two crura (legs), the superior-posterior crus
3
1 General introduction
(leg) of the antihelix and the inferior-anterior crus of the antihelix. The area
between the two crura is known as the triangular fossa. The curved area
that separates the helix and the antihelix is called the scapha. The concha
is the region surrounded by the antihelix. It is divided into two parts by
the curs helix; the superior one is the cymba concha and the inferior part is
the cavum concha. The antitragus is a cartilaginous protrusion situated in
the region that is inferior to the antihelix and superior to the earlobe. The
targus is a small prominence of skin-covered cartilage, anterior to the EC.
At the inferoposterior margin of the targus, the incisura takes its place.
1.1.2 Middle ear
Figure 1.3 shows the ME, or the tympanic cavity, which is the second part
of the ear. The ME is bounded by the TM on its lateral side and by the
IE on its medial side. It is filled with an air medium that comes to it from
the nasal part of the pharynx via the auditory tube. It accommodates the
ossicular chain that covers the gap between the TM and the IE, cf. figure
1.3. The tympanic cavity consists of two parts: the tympanic cavity proper
and the attic or epitympanic recess. The attic contains the superior half of
the malleus and most of the incus body.
In humans, the tympanic cavity has a diameter of about 15 mm in the ver-
tical and antero-posterior directions, including the attic. Beside its main
components (TM, malleus, incus and stapes), the ME has other parts, such
as suspensory ligaments and attaching muscle tendons, and Eustachian tube
(ET). In most mammals, the ME cavity is extended from the EC by a thin
bone, called the tympanic bulla, while in humans it somewhat different. The
tympanic cavity is extended by the surrounding mastoid bone that contains
many interconnected air cell pockets.
The ME is a biomechanical system that is closed off most of time from
ambient pressure by the TM and the ET, and which has a rather constant
gas volume (Dirckx et al., 2008), cf. figure 1.1. The ET (approximately 35
mm long in an adult human) extends from the anterior wall of the ME to
the back of the throat and the nasal cavity. A portion of the tube (1/3) is
made of bone; the rest is composed of cartilage (Kemaloglu et al., 2000).
4
1.1 Ear anatomy
Figure 1.3: Anatomy of the ME (edited from http://www.
familydoctor.co.uk).
5
1 General introduction
The tympanic membrane
The thin TM separates the EC from the tympanic cavity, forming an angle
of about 55
o
to the inferior part of the EC. It is slightly oval shaped with a
long and a short diameter of respectively 9-10 mm and 8-9 mm, for humans.
The TM has a 3D shape of a tent because it has been pulled medially by
the manubrium of the malleus ossicle, which is partially or fully connected
with the TM depending on the species. For instance in humans, the malleus
is connected to the TM at its tip (umbo) and the short process. The TM
shape of different species is presented in figure 1.4.
Figure 1.4: Outlines of TMs of different species (from Decraemer
and Funnell (2008)).
The TM consists of two parts; pars tensa (PT) and pars flaccida (PF). The
PT is divided into four quadrants: the postero-superior, the postero-inferior,
the antero-inferior, and the antero-superior quadrant cf. figure 1.5.
6
1.1 Ear anatomy
Figure 1.5: The human TM (from Gray (1918)).
The PT is composed of three layers: a cutaneous layer (skin) at the lateral
side, a mucous layer at the medial side and a fibrous layer in between. The
cutaneous layer extends from the skin in the EC. The PF, which is about
10 times smaller than the PT in humans TMs, lacks the fibrous layer that
is present in the PT. The PF is relatively thick and relaxed, while the PT is
thin and tense (Kuypers et al., 2006). Though the TM seems very thin, it
is strong enough to stay undamaged against static pressure differences up to
17 kPa
2
(Cameron et al., 1993).
The ossicular chain
The ossicular chain is composed of three bones, which are the smallest present
in a mammalian body. The first ossicle in the chain is the malleus (hammer),
the second one is the incus (anvil) and the last one is the stapes (stirrup).
The ossicular chain physically bridges the gap between the TM (with the
2
The pascal (Pa) is the SI derived unit of pressure that represents the force per unit
area. It is defined as one newton per square metre (1 Pa =
N
m
2
).
Bar = 1×10
5
Pa
Standard atmosphere (atm) 1.013×10
5
Pa
mmH
2
O 9.807 Pa.
7
1 General introduction
malleus) and the oval window of the IE (with the stapes). The three bones
are connected together by articular joints, which are more fixed or loose
depending on species ligaments, as well as muscles (Suwannahoy et al., 2012).
Together, they restrict and dictate the movement of the ossicular chain.
(a) malleus
(b) incus (c) stapes
Figure 1.6: Anatomical details of (a) malleus (b) incus and (b)
stapes (edited from http://www.theodora.com).
The malleus
The malleus, which is named from its fancied resemblance to a hammer,
is the first and largest ossicle in the auditory path of sound. Directly
behind the TM (from the medial side). The malleus consists of a head,
a neck and three processes: the anterior process, the lateral process
and the manubrium, cf. figure 1.6(a). At its manubrium (handle), the
8
1.1 Ear anatomy
malleus is connected to the TM, this connection varies between species,
cf. figure 1.4. Through three ligaments, the malleus is connected to the
lateral side of the tympanic cavity. The anterior ligament is connected
to the neck right above the anterior process. The superior and lateral
ligament both connect to the head of the malleus from the superior
and lateral side respectively. Additionally, the malleus is connected
with one muscle, the tensor tympani, which attaches itself near the
root of the manubrium.
The incus
Figure 1.6(b) shows the incus ossicle that gets its name due to its anvil-
like shape. It consists of a body and two crura. The body is somewhat
cubical but compressed transversely. The anterior surface of the incus
body has a convex-shaped facet where the head of the malleus makes
contact with it. The two crura, they are different in length, are named
as the short crus and the long crus. The short crus (short process) is
somewhat conical in shape, projects almost posteriorly from the body
to the fossa incudis, a small cavity in the TB to which it is attached.
The long crus (long process) dives almost inferiorly from the body and
parallel to the malleus manubrium. The long crus ends at a rounded
projection tip covered with cartilage called the lenticular process where
the incus touches the head of the stapes. While it does not attach to a
muscle, like the malleus, the incus is connected with only one ligament
to the lateral side of the tympanic cavity, which is the superior ligament.
The stapes
The Stapes, which is named from the similarity of its shape to that of
a stirrup, cf. figure 1.6(c), is the smallest bone in the human body. It
consists of a head, a neck, two crus and a base. The head, which is
covered by cartilage, points laterally to make contact with the lenticular
process of the incus. Medial to the head, the neck is situated. The two
crura, the crus anterius and the crus posterius, diverge from the neck
to the medial side and are connected at their end by a flattened oval
plate; the base (basis stapedis). This forms the footplate of the stapes
by which the stapes fits into the oval window of the cochlea. The
footplate is surrounded by the annular ligament. Of the two crura, the
anterior is shorter and less curved than the posterior. The stapes is
attached at the posterior side of its neck to a muscle, the stapedius
muscle.
Morphological discussion about the ME is elaborated in chapter 2.
9
1 General introduction
1.1.3 Inner ear
The IE or labyrinth, figure 1.1, is the third part of hearing organ, which
occupies the most medial side of the ear. It consists of two major parts: the
bony labyrinth and the membranous labyrinth. The first could be further di-
vided into three parts: the vestibule, the semicircular canals and the cochlea,
which are hollowed cavities of a bony structure lined by periosteum, cf. fig-
ure 1.7(a). They contain perilymph, a fluid rich in sodium ions, in which the
membranous labyrinth is situated. The vestibule, where the stapes is firmly
attached to the oval window (fenestra vestibuli), occupies the central part of
the bony labyrinth, medial to the tympanic cavity.
(a) Bony labyrinth (b) Membranous labyrinth
Figure 1.7: Anatomical details of human IE; (a) The bony
labyrinth (b) The membranous labyrinth (from Gray (1918)).
The bony semicircular canals are three in number: superior, posterior and
lateral semicircular canal, situated superiorly and posteriorly to the vestibule.
They have a circular shape, compressed from side to side, with a diameter of
about 0.8 mm but they differ in length, 12 to 15 mm for the lateral, 15 to 20
mm for the superior and 18 to 22 mm for the posterior semicircular canal.
The three canals are perpendicular to each.
The cochlea, which has a spiral shape, forms the anterior part of the labyrinth,
cf. figure 1.7(a). In humans, it measures about 5 mm from base to apex.
The cochlea consists of three chambers or scalae: scala tympani, scala media
and scala vestibuli, cf. 1.8(a). Membranes separate them from each other.
The scala tympani extends from the round window to the helicotrema where
the scala vestibuli starts. The two scalae are separated by the basilar mem-
brane and they contain the same fluid, the perilymph. The scala vestibuli
10
1.1 Ear anatomy
ends at the vestibule. It receives sound vibration from the stapes footplate
and passes it to the scala media. Reissner’s membrane separates the scala
vestibuli from the scala media. The scala media lies between the scala tym-
pani and the scala vestibuli. It accommodates the core component of the
cochlea, the organ of Corti or spiral organ. The latter contains auditory
sensory cells, the hair cells, cf. figure 1.8(b). In humans, the organ of Corti
contains 15,000 to 20,000 hair cells.
(a) Cochlea (b) Corti
Figure 1.8: Cross section of the (a) Cochlea (b) Organ of Corti
(edited from Wiki).
Figure 1.7(b) shows the membranous labyrinth that is lodged within the
bony labyrinth and fixed to it in certain places. The membranous labyrinth
contains endolymph. At its wall, the acoustic nerves are distributed. The
membrane consists of two sacs, the utricle and the saccule. The utricle is
larger than the saccule. It forms the superior-posterior part of the vestibule.
11
1 General introduction
The saccule occupies the region near the opening of the scala vestibuli of the
cochlea.
1.2 Ear function
From the previous section, we see that the ear has many components. One
could ask a logical question; Why does it have such a number of components?
Another could ask, are they all involved in operating and enhancing our
hearing sense and how? In what follows, a physiological overview of the ear
is presented to answer these questions.
The OE has two main functions: it helps localize the source of sound, and
collects and amplifies the acoustic pressure. Sound needs to be amplified
since it passes from air, in the OE, to fluid in the IE, which reflects some
acoustic energy. The OE alters the amplitude of the incoming sound wave
and provides a mechanism to differentially amplify sounds within the audible
frequency range (cf. section 1.3). The OE pinna amplifies sound by a factor
of 2 or 3, which gives a gain of about 6 dB, and directs it into the EC. Some
animals increase the amplification more than that by turning and shaping
their auricles since they have mobile pinnae.
The resonance frequency of the concha and the EC around 5 kHz and the
reflection of sound from the torso together with diffraction of sound by the
head lead to additional amplification of about 5 to 20 dB. At higher frequen-
cies, above 6 kHz, the pinna allows sound frequencies to enter the EC with
a slight delay that is translated into a phase cancellation, which is known as
the pinna notch filter.
The ME has a major influence in the amplification process, however its func-
tion is more complicated than just an amplifier or a puffer. Generally, ME
function can be described by three mechanisms: area ratio, lever action and
“buckling” effect. Since sound is collected by the OE and passed to the ME,
which ends up at the stapes, the sound pressure is concentrated due to tran-
sition from big to relatively small area; this is called the area ratio, cf. figure
1.9(a). Due to the difference between the area of the TM and the stapes
footplate (oval window), the first and largest gain is obtained. In humans
and cats the TM is approximately 17 times larger than the oval window,
which corresponds with an amplification of around 25 dB, while in gerbil
the ratio rises up to 25 times, which corresponds to 28 dB. The second
mechanism is the buckling effect, which originates from the tent-like shape
12
1.2 Ear function
of the TM, cf. figure 1.9(b). Due to its shape, relatively big forces on the
TM result in small deformations, these deformations and associated forces
are concentrated at its centre, where the malleus handle is attached to it.
This effect allows an increase of the force at stapes by a factor of 2, which
corresponds to an amplification of 6 dB (Tonndorf and Khanna, 1970).
At low frequencies, the ossicles also act as a lever, where they move as a single
rigid body rotating around a rotation axis at the incudomalear complex.
Figure 1.9(c) shows that malleus length is longer than stapes length. From
a mechanical point of view, this length difference serves to increase the force
of the TM applied on the stapes footplate. In humans, the lever ratio value
is 1.3, which gives an amplification of 2 dB. The ossicular chain motion
becomes more complex at higher frequencies (Decraemer et al., 1989), thus
amplification factor due to lever action becomes less straightforward.
The ET, as a component of the ME, has a special function. Opening and
closing of the ET are physiologically and pathologically important. Normal
opening of the ET equalizes atmospheric pressure in the ME; closing of the
ET protects the ME from unwanted pressure fluctuations and loud sounds.
Mucociliary clearance drains mucus away from the ME into the nasal part of
the pharynx, thus preventing infection from ascending to the ME is another
function of the ET (Bluestone and Doyle, 1988; Bluestone, 1996). We clearly
feel that we need the opening of the Eustachian to equalize the sudden pres-
sure change when we, for instance, take an elevator, dive in water or during
take-off/landing on an aeroplane (Dirckx et al., 2006).
We arrive now at the IE and the cochlea; which is considered to be the
heart of the hearing organ. The cochlea contains fluid, which has a different
resistance to sound than in air. The resistance of sound transport through a
medium is called acoustic impedance Z, which is defined as the ratio of sound
pressure p and the particle velocity v. It is measured in rayl unit (N.s/m),
cf. equation 1.1.
Z =
p
v
. (1.1)
When sound energy passes the border between two media with different
acoustic impedance, it has three possibilities: reflection, absorption, or trans-
mission, a combination of two is also possible. The amount of reflected acous-
tic energy R depends on the impedance difference between the two media,
equation 1.2.
13
1 General introduction
(a) Area ratio (b) Buckling effect
(c) Lever action
Figure 1.9: The ME amplification mechanisms: (a) Area ration;
the difference between the area of TM and the area of oval window
(stapes footplate), from Gelfand (2009). (b) Buckling effect of the
TM, from Tonndorf and Khanna (1970) and (c) Lever action. A
TM
:
the area of the TM, A
s
: oval window area (stapes footplate), I
m
:
malleus length and I
i
: incus length. The left image is obtained
from a µCT morphological model of a gerbil, cf. chapter 2.
14
1.3 Sound pressure regimes
R =
Z
2
Z
1
Z
2
+ Z
1
!
2
. (1.2)
From equation 1.2, when sound comes from air (Z
1
400 rayl) at the OE to
the cochlea which contains fluid (Z
2
1.5 x 10
6
rayl), we get that 99.9% of
the incoming acoustic power is reflected, which leads to loss in the acoustic
energy of 60 dB. The energy loss by the reflection is approximately equal to
the amplification of the acoustic energy that is made by the OE and mainly
by the ME as we have seen earlier in this section.
Finally, the hair cells are responsible for transforming sound into electrical
signals that nerves can convey to the brain, to be interpreted, with the help
of charged particles (primarily potassium and calcium). Moreover, hair cells
play a key role in human balance. The hair cells that are located in the
vestibular (balance) organs of the IE detect changes in head position and
send this information to the brain via nerve fibers. This information is used
to help maintain body posture, eye position and balance.
1.3 Sound pressure regimes
If someone asks what we hear, the obvious answer is “sound”, the question
then becomes what is sound? But simply, sound is a pressure wave, which
consists of pulsations or vibrations of molecules of an elastic medium such as
gas, liquid or solid (Manhart, 1974). For the hearing process, sound exists in
acoustical, mechanical, hydraulic and electrical forms. The acoustic pressure
(sound pressure) is collected by the OE passed to the ME to be converted
into mechanical energy which is then transformed into hydraulic energy in
the endolymph of the IE and finally to electrochemical energy by the hair
cells to allow sound travel through the nerve to the brain for interpretation.
The sound pressure is presented as a logarithmic measure of the effective
sound pressure of a sound relative to a reference value (p
ref
) of 20 µPa that is
usually considered as the threshold of human hearing (at 1 kHz), measured
in decibels (dB).
dB
SPL
= 20log
10
p
p
ref
!
. (1.3)
For humans, the sound pressure level has a wide range, spread over seven
15
1 General introduction
orders of magnitude starting from 0 dB (the threshold of hearing at 1 kHz)
to 140 dB (pain threshold at 130 dB), cf. figure 1.10 and table 1.1.
Figure 1.10: Schematic diagram represents approximately the
frequency and pressure spectrum and the range of human percep-
tion (edited from Gea et al. (2010)).
Some examples of the sound pressure levels of common sound sources in our
environment are presented in table 1.1.
The sound pressure is not the only factor that controls what we hear; another
important factor is the frequency of sound. It is well known that we can only
hear a limited frequency range. This limitation is called the audible range or
hearing range, which has been a subject of study since the middle of the 19
th
century. Galton (1883) studied the response of animals to high-frequency
sounds. He reported significant species differences in the ability to hear high
frequencies. Among the animals he studied, cats were found to have the best
high-frequency hearing. He claimed that cats need to hear the high frequency
sounds made by their prey such as mice and other small animals (Galton,
1883).
Since then it has been clear that there is a variation in the hearing ability
among animals. Many more recent studies have been carried out to determine
the hearing range for animals (Rosenzweig et al., 1955; McCormick et al.,
1970; Kreithen and Quine, 1979; Heffner and Masterton, 1980; Heffner and
Heffner, 1985; Fay, 1988; Heffner and Heffner, 2007, 2008).
In humans, the audible range of frequencies is 20 Hz to 20 kHz, nevertheless
there is a considerable variation between individuals especially at the high
16
1.3 Sound pressure regimes
Table 1.1: Examples of sound pressure levels in dB and corre-
sponding sound pressure in Pa of some sound sources.
Source sound pressure
level in dB
corresponding
pressure in Pa
Jet aircraft, 50 m away 140 200
Threshold of pain 130 63.2
Threshold of discomfort 120 20
Chainsaw, 1 m distance 10 6.3
Disco concert, 1 m from speaker 100 2
Diesel truck, 10 m away 90 0.63
Kerbside of busy road, 5 m 80 0.2
Vacuum cleaner, 1 m away 70 0.063
Conversational speech, 1 m distance 60 0.02
Average home 50 0.0063
Quiet library 40 0.002
Quiet bedroom at night 30 0.00063
Background in TV studio 20 0.0002
Calm breathing 10 0.000063
Threshold of hearing 0 0.00002
17
1 General introduction
frequencies, where a gradual decline with age is considered normal. The
hearing range for some animals and humans is illustrated in figure 1.11.
Figure 1.11: The hearing range of different mammals. Thin
lines indicate the range of frequencies that can be detected at 60
dB SPL; thick lines indicate the range of frequencies that can be
detected at 10 dB SPL, which represents the range of good hearing
(edited from Heffner and Heffner (2007)).
From figure 1.11, it is clear that most mammals have a better high-frequency
hearing limits (upper frequency limit at 60 dB SPL) than humans do. That
ranges from 34.5 kHz for Japanese macaque to 85.5 kHz for the domestic
house mouse. The main reason for this variation is attributed to the fact that
small mammals need to hear higher frequencies than do larger mammals in
order to make use of the high-frequency sound localization cues provided by
the attenuating effect of the head and pinnae on sound (Heffner and Heffner,
2007).
18
1.3 Sound pressure regimes
For the low frequency limit, humans seem to have better lower limit than
other mammals do. At 60 dB SPL, the range of lower limit is 28 Hz for the
Japanese macaque, which slightly exceeds that of humans of 31 Hz, to 2.3 kHz
for the domestic mouse. Only Indian elephant is known to have significantly
better low-frequency hearing (17 Hz) than humans. Although the reason for
the variation in the lower limit is not fully understood, Heffner and Heffner
(2003) attributed it to the fact that some animals have reduced their low-
frequency hearing to prevent the low-frequency components of sounds from
masking the high-frequency components they need for sound localization
As shown in figure 1.10, the spectrum of frequencies and pressures that the
humans face at normal circumstances could be divided in three pressure
regimes: static, quasi-static and acoustic. The quasi-static pressure regime
marks the transition between static and acoustic changes, but its limits are
arbitrary. The effect of both static and acoustic pressures on hearing were
investigated in many studies to point out the motion of TM and the ME
ossicles as well as sound transfer to the cochlea (Tonndorf and Khanna, 1970;
McPherson et al., 1976; Buunen and Vlaming, 1981; von Unge et al., 1991;
Dirckx and Decraemer, 1992; Decraemer et al., 1994; Lee and Rosowski, 2001;
Dirckx, 2006; Mills et al., 2007; Dai et al., 2008; Chien et al., 2009; Gaihede
et al., 2010; Cheng et al., 2010; Koka et al., 2011).
Apart from acoustic pressures, the ME is also subject to very slow pressure
variations due to, for instance, gas exchange processes through the ME mu-
cosa, which acts as a natural pumping mechanism. Moreover, we pressurize
our ears with thousands of Pascal when we take off with an aeroplane, dive
in the water or take an elevator for a couple of floors. These quasi-static
pressure variations have larger amplitudes than the highest sound pressures.
Recently, a number of studies were published (Tideholm et al., 1998; Dirckx
et al., 2006, 2008) concerning 24-hr. real-time middle ear pressure (MEP)
monitoring on humans. They reported that MEP could change up to 1 kPa
within minutes in our daily activities. Dirckx et al. (2008) demonstrated that
the pressure rose about 0.5 kPa in a time span of just 30 seconds, when an
elevator is used.
Changes were also recorded when a patient lay down on a bed or got up again.
When going to a recumbent position, MEP increased by 1.3 kPa in a time
span of one hour, while the pressure dropped quickly when s/he went back
into an upright position. Even blowing the nose pressurises the ME in seconds
with about 100 Pa. The associated TM and ME ossicles motion with the
quasi-static pressure changes have not been thoroughly investigated, despite
its importance in understanding how the ear works at such high pressures.
19
1 General introduction
Few studies reported the ME mechanics in the static and quasi-static pressure
changes such as Hüttenbrink (1988); Dirckx et al. (2006); Gea et al. (2010).
Hüttenbrink (1988) studied the displacement of malleus, incus and stapes at
pressures equivalent to 0.5, 1, 1.5, 2 and 4 kPa using microscopy. These mea-
surements were also done by Dirckx et al. (2006) on the umbo and stapes of
the rabbit ME using laser Doppler vibrometry (LDV). With pressure change
rates of 0.2 kPa/s (20 mHz) to 1.5 kPa/s (150 mHz), displacement as at a
pressure up to ± 2.5 kPa was measured.
Displacements in the human and gerbil MEs at different static pressures have
also been measured by Gea et al. (2010). At static pressures of ± 5 kPa, the
displacements were calculated from data obtained with µ-CT scans. The
µ-CT scan setup offers some 3D information about the motion, but at static
pressure changes. In this thesis, the mechanics of the ME in the quasi-static
pressure regime with dynamic pressure changes (0.5 to 50 Hz) is presented.
1.4 Motivation
1.4.1 Hearing loss
Recently, the WHO has ranked hearing loss as one of the six most common
diseases affecting the lives of people in industrialized countries. For children,
hearing loss can change their live dramatically backward; as one sees that
people with hearing and communication problems are always try to avoid in-
teraction with others that is in return depriving them of acquiring knowledge
from their peers. Moreover, if no serious care is taken, educational process
may be affected and thus the future of those people becomes grey, if not
dark.
Young people (and adults) are in danger too due to their attitude in the
leisure time; because they are often exposed to very loud sound that some-
times reaches the pain threshold, such as in big concerts. Nowadays, it is
common to see people wearing earphones and listening to loud music for sev-
eral hours on a daily basis. Recently, Barbara et al. reported that “stress”
must be taken into account as an additional factor that has an influence on
the hearing problems (Barbara et al., 2012). They concluded that long-term
stress and other Psychiatric disorders in the western countries, as in Sweden
where the study took place, may lead to increase the number of the affected
people. In addition, noise pollution in big cities becomes unavoidable. As-
20
1.4 Motivation
sociated with other factors, therefore the figure for people with hearing loss
increases to be a true problem. Accordingly, WHO estimated this number to
be 278 Million worldwide in 2005, which is about 5 % of the world population
at that time.
Hearing loss, also known as hearing impairment, is defined as any degree
or type of auditory disorder, while deafness is known as the inability to
discriminate conversational speech. One is considered as deaf if s/he needs
to receive sound of at least 90 dB in order to hear it. This shows that even
people who can hear above that loudness are considered as deaf.
Types
Hearing loss is sub-divided into two major types due to the causes:
Conductive hearing loss: This occurs when sound cannot be conducted
from the OE to the IE due to the OE and/or ME dysfunction. The
major reason for this kind of hearing loss for children and youths is
chronic otitis media, which occurs when the ME pressure cannot be
equalized due to malfunction of the ET. With built up pressure, the
environment becomes suitable for bacteria to colonize the ME, thus
inflammation which leads to distort and perforate the ME. Some of
other reasons for conductive hearing loss are ruptures or holes in the
TM, fixation of the ME ossicles, exposing to a sudden high pressure
change, head injury, severe repeated infection and tumours.
Sensorineural hearing loss; It happens due to any kind of damage in
the nervous system that is involved in hearing such as the IE, auditory
nerve, brain stem and cerebral cortex, where sound is interpreted. The
main reason for this kind of hearing loss is abnormalities in the hair
cells. This kind of disorder can be mild, moderate, or severe. The
latter case leading to total deafness. The Major causes for sensorineural
hearing loss are genetic defects, which deny the proper development of
the IE, as well as exposing to complications during pregnancy, birth and
childhood. This includes taking some medications, head injuries and
exposure to sudden high sound pressures. Nevertheless diminishing of
hearing ability with time and exposure to noise pollution are considered
as the major reasons for hearing loss. Accordingly, 40% of people older
than 65 years are affected.
When someone has both above mentioned sorts, a third type of hearing
loss becomes visible. This combined disorder is less common than both
21
1 General introduction
conductive and sensorineural hearing loss individually.
Treatment
The treatment of the hearing loss can be performed in several ways according
to the cause and level of it. The easiest and most common way is the hearing
aid, which helps more than 90% of people with hearing problems. It works
by amplifing the acoustic sound to a level that can be heard. There are
numerous types of hearing aid, such as behind the ear aids or in-the-ear aids,
cf. 1.12.
Dispite its easyness and availabilty, hearing aid is not the ideal method of
treatment as they come with many drawbacks, e.g. uncomfortable, back-
ground noise and social stigma that is attached to hearing loss, which tragi-
cally still exists.
Figure 1.12: Illustration of different hearing aid types (edited
from Wiki).
The treatment of conductive hearing loss can also be achieved by reconstruc-
tive surgery (Marquet, 1971; Gea et al., 2010) of the TM, which is known as
22
1.4 Motivation
tympanoplasty or/and ME ossicles, ossiculoplasty. With ossiculoplasty, the
ME ossicles are replaced by ossicles from a human donor (complete ossicular
chain or in part). It needs that the patient should find a donor fits him. As
this seems difficult due to the availability of a suitable donor, prostheses (im-
plants), made of titanium, becomes a good alternative. As shown in figure
1.13, the ME ossicles can be replaced entirely, right image, or just for the
affected ones.
Complications for this approach come from the size of the ME; which makes
it difficult for surgeons. Moreover, the human body frequently rejects foreign
body material whether it is from a donor or implants.
Figure 1.13: Illustration of different types of the ME implants
(from Gea et al. (2010)).
When severe hearing loss is diagnosed, especially for the combined hearing
loss, cochlear implants can be an effective way to treat damage of both the
ME and IN. As illustrated in figure 1.14, cochlear implants pick up and
process sound waves and then transmit electrical pulses to the acoustic nerve.
Since the 1950s, efforts have been performed to improve the quality and
availability of cochlear implants. Consequently, they have benefited more
than 250,000 people worldwide. Nevertheless they are still expensive and
need appropriate centres and expert surgeons to be implemented. Moreover
it is possible that cochlear implants can be rejected by the human body, since
they are foreign materials.
1.4.2 Hearing research impact
Due to its importance, the ear, thus the hearing sense in general, has been
a subject of research for centuries. In the 16
th
century, Bartolomeo Eustachi
(1500-1574), Giovanni Filippo Ingrassia (1510-1580), Andreas Vesalius (1514-
1564) and Costanzo Varolio (1543-1575) reported the anatomical structure of
the ME. Later on, in the 19
th
century, Adam Politzer (1835-1920), who is one
23
1 General introduction
Figure 1.14: An example of a cochlear implant: (1) A small
microphone that captures sound wave and converts it into dig-
ital code, (2) A receiver and processor unit that transmits the
digitally-coded sound through the coil to the implant, (3) An im-
plant to convert sound into electrical impulses and sends them
along the electrode array, which is positioned in the cochlea and
(4) Electrodes to stimulate the cochlea’s hearing nerve, which
then sends the impulses to the brain. (edited from http://www.
cochlear.com).
24
1.4 Motivation
of the pioneers and founders of otology, opened the doors widely for the those
following to build upon his valuable contributions on ossicles movement.
In the early 1870s, Ernst Mach (1838-1916) and Johann Kessel (unknown)
have observed and reported the deformations of the TM in the static-pressure
regime (Hui, 2008). A few years later, Hermann von Helmholtz (1821-1894)
published his curved-membrane theory which was the first theory to report
the lever system of which the TM is involved (Helmholtz, 1868).
Since then, so many researchers have put their hands to push the wheel
toward better understanding of hearing and thus helping people who suffer
from hearing difficulties. Among them, von Békésy, Funnel, Tonndorf and
Khanna who have made important contributions that are highly appreciated.
Generally, the contributions in the field can be can be divided into several
interactive directions (cf. figure 1.15) that facilitate the hearing research.
For anatomy and geometry of the ear, morphological computer models at-
tracted some attention in order to provide accurate estimation of the ear’s in-
ternal structures, which is important for the prostheses as well as for the finite
element (computer) modelling (FEM). The movement and the biomechanics
of the ME has recently been studied by, e.g., Hüttenbrink (1988); Hergils
et al. (1990); von Unge et al. (1993); Decraemer et al. (1994); Rosowski et al.
(1999); Dirckx et al. (2006); Chien et al. (2007); Aerts and Dirckx (2010);
Cheng et al. (2010); Gea (2010); Volandri et al. (2011); Aernouts et al. (2012).
The FEM has become an established numerical technique to simulate ME
mechanics. In ME research, the technique was first introduced by Funnell
and Laszlo (1978). Since then, so many studies have been carried out in
order to better understand the complexity of the ear and hearing and thus
helping to ease the treatment procedure (Williams and Lesser, 1990; Wada
et al., 1992; Prendergast et al., 1999; Eiber et al., 1999; Koike et al., 2002;
Sun et al., 2002; Gan et al., 2004a; Wang et al., 2007).
Hearing aids and prosthesis benefit from these developments as better under-
standing of ear anatomy and physiology is led to more effective treatment.
An effective prosthesis highly depends on its mechanical properties such as
shape and weight. These have to be tested before implementation, and trial
and error tests in human volunteers is the most well known available method.
With the availability of computer models, surgeons can model the ear with
prosthesis and calculate the necessary parameters to optimize the final re-
sults. Therefore, it reduces the cost, pain and waiting time to have well-fitted
prosthesis to patients. Moreover, computer modelling facilitates the training
of new surgeons to have enough knowledge to perform surgeries.
25
1 General introduction
Figure 1.15: Schematic drawing representing hearing research
and its interactive input/output
The current work comes as a continuation of the previous studies to con-
tribute to the field. We noticed that the available morphological computer
models of the ME lack some important features due to their low resolution.
It is well known that the FEM is highly dependant on the accurate data of
the shape, boundaries, location and dimensions of the ME and IE structures.
Therefore, we create and present high-resolution 3D morphology models for
different species based on a state-of-the-art µCT scanner. With these models
we predict that modellers can achieve more precise results and thus it eases
the life for those who suffer from hearing troubles.
As seen in this thesis, human ears are subject to different pressure changes,
both static and dynamic. Though a lot of effort has been made to study
the associated effects at different pressure regimes, the effects of quasi-static
pressure changes on the ear and hearing are still not entirely clear. Therefore,
this work presents a new approach with micro-scale accuracy to study the
3D micro motion of the ME in the quasi-static pressure regime. The output
of this study can open the doors for those following to fully understand the
3D motion of the ME ossicles at different pressure changes which leads to
better and more realistic prosthesis and hearing aids.
26
Chapter 2
High resolution 3D morphological
models of the middle ear
abstract
Middle ear (ME) finite-element (FE) models are an important tool in hearing
research to improve the understanding of this complex mechanical system. In
order to improve realism in ME finite-element modelling (FEM), comprehensive
and precise morphological data are needed. Therefore, this study was conducted
to present high-resolution 3D morphology models of cat, gerbil, rabbit, rat and
human ossicular chains. The models are based on high-resolution CT measure-
ments. The resolution of the CT images, from which the models are segmented,
varies from 5.6 to 33.5 µm depending on the size of the sample. These high-
resolution models can be used to generate precise FEM of the ME. In addition,
the gerbil and rabbit models are used in this dissertation to demonstrate ossicles
motion, which is the main goal of my Doctoral research.
This chapter is based on:
J.A.N. Buytaert, W.H.M Salih, M. Dierick, P. Jacobs and J.J.J. Dirckx. Realistic 3-D model of the
gerbil middle ear, featuring accurate morphology of bone and soft tissue structures. Journal of the Asso-
ciation Research in Otolaryngology (JARO), 12:681–96, 2011.
Open-access availability of these models was announced as short communication paper in the journal
hearing research:
W.H.M Salih, J.A.N. Buytaert, J.R.M. Aerts, P. Vanderniepen, M. Dierick and J.J.J. Dirckx. Open
access high resolution 3D models of cat, gerbil, rabbit, rat and human middle ears. Hearing Research,
284 (1-2): 1–5, 2012.
27
2 High resolution 3D models
2.1 Introduction
Since it has been introduced by Funnell and Laszlo (1978), the FEM at-
tracted some attention to study the complex mechanics of the ME. As one
of its inputs, FEM requires 3D morphological computer models of the ME
components. These mesh models consist of a finite number of elements, e.g.,
tetrahedra or hexahedra.
Current morphological models are either incomplete, low resolution, and/or
contain rudimentary shapes to represent (some) ME components. Since the
criteria of low, modest or high resolution have no fixed definition, we divided
previous work as follows: We talk of low resolution when the models look
coarse, pixelated, roughly triangulated and/or only contain simple structures.
Modest resolution covers model shapes which look more smooth and natural,
though still lack small detailed features resolved in our models. Recently,
a new term is introduced to distinguish between the absolute resolution in
µm and the relative resolution which is given by dividing the voxel size by
the sample size. When this ratio is on the order of 1/250, it is considered
as a low relative resolution, while the high resolution is achieved when the
ratio becomes on the order of 1/2000. Pioneering work in this field used
manually drawn geometrical shapes in the computer to represent the ME
malleus, incus, and stapes (Wada et al., 1992; Ladak and Funnell, 1996;
Blayney et al., 1997; Prendergast et al., 1999; Eiber et al., 2000; Koike et al.,
2002).
Some authors used low- or modest-resolution shapes measured with medi-
cal CT (Rodt et al., 2002; Lee et al., 2006a) or with tabletop µCT devices
(Decraemer et al., 2002, 2003; Elkhouri et al., 2006; Lee et al., 2010; Puria
and Steele, 2010). Other authors used histological sectioning (Funnell et al.,
1992; Sun et al., 2002) or magnetic resonance microscopy (MRM, NMR,
MRI) (Funnell et al., 2005; Elkhouri et al., 2006), but again with modest res-
olutions. In many models, the suspensory ligaments and muscle tendons are
either omitted (Wada et al., 1992; Ladak and Funnell, 1996; Blayney et al.,
1997; Rodt et al., 2002) or manually incorporated as simple geometrical ob-
jects such as blocks, cylinders, or cones (Prendergast et al., 1999; Beer et al.,
1999; Koike et al., 2002; Sun et al., 2002; Lee et al., 2006b).
As of today, only models by Wang et al. (2006); Gan et al. (2007), and Cheng
and Gan (2008) (using histological sectioning) and by Mikhael et al. (2004);
Sim and Puria (2008), and Ruf et al. (2009) (using x-ray techniques) contain
actual measured shapes of soft tissue structures, but in low resolution.
28