COCHLEAR IMPLANTS. Fundamentals and Applications

Transcription

1 COCHLEAR IMPLANTS

2 COCHLEAR IMPLANTS Fundamentals and Applications Graeme Clark The University of Melbourne and The Bionic Ear Institute, East Melbourne, Victoria, Australia With 244 Illustrations A I P ' Springer

3 Graeme Clark Department of Otolaryngology and The Bionic Ear Institute The University of Melbourne Albert Street East Melbourne, Victoria 3002 Australia Series Editor: Robert T. Beyer Department of Physics Brown University Providence, RI USA Library of Congress Cataloging-in-Publication Data Clark, Graeme. Cochlear implants : fundamentals and applications / Graeme Clark. p. cm. (Modern acoustics and signal processing) Includes bibliographical references and index. ISBN (alk. paper) 1. Cochlear implants. 2. Deaf Rehabilitation. I. Title. II. AIP series in modern acoustics and signal processing. RF305.C dc ISBN Printed on acid-free paper Springer-Verlag New York, Inc. AIP Press is an imprint of Springer-Verlag New York, Inc. All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Springer-Verlag New York, Inc., 175 Fifth Avenue, New York, NY 10010, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. Printed in the United States of America SPIN Springer-Verlag New York Berlin Heidelberg A member of BertelsmannSpringer Science Business Media GmbH

4 This book is dedicated to my wife, Margaret, for her selfless support and wise counsel during the last 35 years while I was undertaking cochlear implant research at the University of Sydney, the University of Melbourne, and The Bionic Ear Institute. I would like also to express my appreciation to our children Sonya, Cecily, Roslyn, Merran, and Jonathan; their spouses Ian, Peter, and Marissa; and our grandchildren Elise, Monty, Daniel, Noah, and Rebekah for their encouragement and enriching our lives.

5 I have been very impressed by the emergence of the bionic ear as a practical proposition, but even more by the promise for the future that it seems to embody. It makes use of the arrangement in the cochlea for pitch recognition to bring electronic technology into direct functional relationship with the nervous system and the human consciousness. Maybe that unique relationship has no other parallel in the nervous system, and thus that direct link between electronics and physiology will find no other application to medicine. Nevertheless, I feel it may represent a new benchmark in the understanding of neural and mental function in terms of their physical components. Professor Emeritus Sir Macfarlane Burnett, A.K., O.M., K.B.E., M.D., Ph.D., Lond., F.A.A., F.R.S., Nobel Laureate (Physiology or Medicine) The First Patron of the Bionic Ear Institute, 1985

6 Series Preface Soun is nought but air y-broke Geoffrey Chaucer end of the 14th century Traditionally, acoustics has formed one of the fundamental branches of physics. In the twentieth century, the field has broadened considerably and become increasingly interdisciplinary. At the present time, specialists in modern acoustics can be encountered not only in physics departments, but also in electrical and mechanical engineering departments, as well as in mathematics, oceanography, and even psychology departments. They work in areas spanning from musical instruments to architecture to problems related to speech perception. Today, six hundred years after Chaucer made his brilliant remark, we recognize that sound and acoustics is a discipline extremely broad in scope, literally covering waves and vibrations in all media at all frequencies and at all intensities. This series of scientific literature, entitled Modern Acoustics and Signal Processing (MASP), covers all areas of today s acoustics as an interdisciplinary field. It offers scientific monographs, graduate-level textbooks, and reference materials in such areas as architectural acoustics, structural sound and vibration, musical acoustics, noise, bioacoustics, physiological and psychological acoustics, speech, ocean acoustics, underwater sound, and acoustical signal processing. Acoustics is primarily a matter of communication. Whether it be speech or music, listening spaces or hearing, signaling in sonar or in ultrasonography, we seek to maximize our ability to convey information and, at the same time, to minimize the effects of noise. Signaling has itself given birth to the field of signal processing, the analysis of all received acoustic information or, indeed, all information in any electronic form. With the extreme importance of acoustics for both modern science and industry in mind, AIP press, now an imprint of Springer- Verlag, initiated this series as a new and promising publishing venture. We hope that this venture will be beneficial to the entire international acoustical community, as represented by the Acoustical Society of America, a founding member of vii

7 viii Series Preface the American Institute of Physics, and other related societies and professional interest groups. It is our hope that scientists and graduate students will find the books in this series useful in their research, teaching, and studies. As James Russell Lowell once wrote, In creating, the only hard thing s to begin. This is such a beginning. Robert T. Beyer Series Editor-in-Chief

8 Preface The cochlear implant is a device that bypasses a nonfunctional inner ear and stimulates the hearing nerves with patterns of electrical currents so that speech and other sounds can be experienced by profoundly deaf people. It is the culmination of investigations that started in the 19th century, and as such it is the first major advance in helping profoundly deaf children to communicate since the sign language of the deaf was developed at the Paris Deaf School 200 years ago. It is also the first direct interface to the central nervous system to restore sensory function for use on a regular clinical basis. I became interested in helping deaf people hear when I was 10 years old, because my father had a severe hearing loss and I knew how difficult it was for him to cope as a pharmacist and as a family man. In 1966 I left my practice as an ear, nose, and throat surgeon in Melbourne to do research and to learn how it might be possible to help people with a profound hearing loss. These were the patients I had to turn away from my clinic, saying that a hearing aid would be of little help but that one day medical research might provide an alternative. For me this meant first undertaking basic studies to learn about the differences between acoustic and electrical stimulation of the auditory neural pathways. When it became clear from these and other basic studies that the best chance of providing speech understanding was through multiple electrode stimulation, many scientific challenges were to lie ahead. As previous attempts to produce speech understanding with electrical stimulation had been unsuccessful, and as reproducing the coding of sound was not seen as feasible, the research faced rigorous scientific criticism. The first criticism came from auditory neuroscience, where research had shown the complexity of the inner ear and central brain pathways. Not surprisingly it was believed that inserting a relatively small number of electrodes into the inner ear to stimulate groups of nerve fibers would fail to produce sufficient information for speech understanding. The second criticism came from the biological and clinical disciplines. Here the concern was that implantation would damage the very nerves it was intended to stimulate. In addition, it was thought the electrode could be a pathway for middle ear infection to induce ix

9 x Preface dangerous infection in the inner ear. These biological and clinical criticisms were also well founded. The delicacy of the inner ear had been appreciated in ear surgery, and the risk of infection was ever present in young children. The above two major criticisms required answers before clinical studies could be done on patients. It was also essential to determine from a small group of volunteers how the complex signals of speech could be presented as patterns of electrical stimulation that could be understood. This seemed at the time an almost insurmountable challenge. Research that followed established that speech processing could in fact be achieved safely for profoundly deaf adults, who had hearing before going deaf. After the benefits were shown for adults, it was appropriate to initiate research to see if children born deaf or deafened early in life could obtain sufficient speech understanding to enable them to manage successfully in a hearing world. Would deaf children be able to develop the right central neural connections, as they had received no auditory stimulation during the plastic phase of brain development? Indeed children who were born deaf were shown to develop speech at a level comparable to that in adults who had prior exposure to sound. Furthermore, it was discovered that if they were operated on at a young age, they could develop good speech sounds as well as language. Providing hearing and speech understanding for children born deaf then led to an intense ethical debate. The signing deaf community had developed an effective communication system and support network to help one another. Community members were upheld by a strong belief in their self-worth, which is so necessary to manage in a world of sound where people with good hearing did not fully appreciate the great difficulty they had. For a time the implant was seen as an ideological threat to their beliefs and as undermining this well-knit group, and for a number of years the efficacy of the procedure was questioned. It required many controlled studies and the opinion of educators who had experience with the achievements of children with hearing aids before the benefits of the implant for children were fully appreciated. The cochlear implant has been the result of research in many disciplines, including surgical anatomy, surgical pathology, biology, biophysics, neurophysiology, psychophysics, speech science, engineering, surgery, audiology, rehabilitation, and education. Few medical advances have required the integration of so many disciplines. The scientific questions in these disciplines had to be addressed in a logical, systematic, and sequential manner, and are discussed in this book. As a result of this research, cochlear implantation has grown from a small number of isolated experimental studies done by a few, to a diverse discipline investigated by many. Its scientific credibility has been recognized through its inclusion in international physiological, acoustical, surgical, otolaryngological, audiological, education, speech science and technology, and engineering society meetings. In addition, there are many international meetings devoted solely to the topic of cochlear implants. The growth in knowledge in the last 30 years has been rapid. This can be seen in the number of papers that include cochlear implants in the title, abstract, or subject heading: in the 1960s, one; in the 1970s, 72; in the

10 Preface xi 1980s, 679; in the 1990s, 1,935. There have been many other relevant publications. Not only have there been a very large number of scientific papers, but there also have been monographs and book chapters. Initially the field drew on basic sciences for its development, and then gradually established its own body of scientific and clinical knowledge. This has continued to the point that now electrical simulation of the auditory system can justly claim to be making scientific contributions to the disciplines that helped establish it, in particular neurophysiology, biology, psychophysics, speech science, and the clinical disciplines of surgery, audiology, and rehabilitation. One aim of this book is to show how the numerous disciplines have contributed and how they have interrelated. This book presents the fundamentals of the research as well as the clinical outcomes so that the reader will have a more complete understanding of the discipline. It is intended for a general reader, and those with a more specialized background can refer to the references. In presenting the fundamentals, research at the University of Melbourne/Bionic Ear Institute and elsewhere is cited. Clinical studies cannot be divorced from the basic research. The two must guide each other and the main aim should be to help people. This requires excellent basic research, but it should be focused and not an end in itself. In this book this interaction is presented at all opportunities. Finally, the basic and clinical research would not have reached the wider community without the biomedical and engineering expertise of industry. The work has been much more demanding than developing a pacemaker, as more complex electronics have had to be encapsulated in a smaller implanted package. Furthermore, the interface with the auditory nervous system is a very intricate bioengineering achievement. For this reason this book not only presents the basic and clinical research, but also discusses how these have supported the industrial achievement. Graeme Clark

11 Acknowledgments Writing this comprehensive work has been a major task, but having been involved with the contributing disciplines since the 1960s has been of great help especially in understanding their evolution and interrelation. Research germane to this book, like the field itself, has been interdisciplinary, and has depended on a coordinated team approach. A team was also required to bring the elements of this book together. I am greatly indebted to the tremendous support received from Sue Davine, David Lawrence, Helen Reid, and John Huigen in the preparation of this book. It has been a major undertaking, and Sue has completed many hours of typing and compiling the text. David has produced diagrams and figures of a very high standard, researched topics, and provided invaluable help. Helen has diligently searched for references and found them, and John has helped to coordinate this combined effort. I would also like to thank in particular Andrew Vandali, Anthony Burkitt, David Grayden, Ian Rutherfurd, Jim Patrick, Joanna Parker, Mark Harrison, Peter Busby, Peter Seligman, Richard Dowell, Thomas Stainsby, and Chris van den Honert for kindly reading sections of the text and for their helpful comments. The imperfections are mine alone. Thanks also to Russell Brooks for compiling the index. The cochlear implant research in Australia has been a team effort, and it has been a privilege to have worked with young and talented research students and to have seen them develop into mature scientists. It has also been a valued experience to have been closely involved with the staff members of Cochlear Limited, a number of whom were research colleagues. Without the close relationship between the basic and focused research in Melbourne and the industrial research and development in Sydney, the Nucleus device would not have become available to tens of thousands of severely and profoundly deaf people in more than 120 countries. I would also like to pay tribute to the many scientists and clinicians from other centers who have also worked with dedication to achieve hearing and speech for deaf people. There has been great collegiality internationally in this relatively small field that has crossed political and other divides to make the world a better place for people with a hearing disability. Finally, the work would not have been possible without the belief and support of many benefactors, governments, trusts, and foundations. Sometimes simply having their encouragement at difficult times was enough. Graeme Clark xiii

12 Contents Series Preface Preface Acknowledgments Abbreviations vii ix xiii xxv Introduction xxxi Definition...xxxi Normal Hearing...xxxi Deafness...xxxi Overall Concept of the Bionic Ear...xxxii Training in the Use of the Bionic Ear...xxxiii Fundamental Objections and Questions...xxxiii Answers to the Fundamental Objections...xxxiv 1. A History 1 Pre-science...1 Eighteenth Century...1 Nineteenth Century...3 Twentieth Century to 1930s: Early Hearing Aids s to 1940s: Initial Indirect Electrical Stimulation in the Human s to 1960s: Initial Direct Electrical Stimulation in the Human s: Fundamental Research in the Experimental Animal s: Fundamental Research in the Experimental Animal and Human s: Fundamental Research, Industrial Development, and Clinical Trials xv

13 xvi Contents 1990s: Continuing Fundamental Research and Industrial Development References Surgical Anatomy 58 Overview Temporal Bone Components Embryology Mastoid Air Cell System and Variations Blood Supply and Innervation Infant and Young Child External Ear Pinna External Auditory Meatus Middle Ear Ossicles Muscles Relationships Posterior Tympanotomy Round Window and Niche Inner Ear Osseous Membranous Histology of the Cochlea Embryology Central auditory system Overview Auditory Nerve Cochlear Nucleus Superior Olivary Complex Lateral Lemniscus Inferior Colliculus Superior Colliculus Medial Geniculate Body Auditory Cortex References Surgical Pathology 100 Inflammation Classification Etiology Pathophysiology Insertion Trauma Tissue Responses in the Cochlea of the Experimental Animal...104

14 Contents xvii Tissue Responses in the Human Bio-compatibility of Materials Methods of Investigation Tissue Response Infection Otitis Media Labyrinthitis and Meningitis Experimental Animal Studies Host Factors and Foreign Bodies Clinical Protocol Deafness and the Central Auditory Pathways Spiral Ganglion Cochlear Nucleus Pons and Midbrain Human Brainstem Prenatal (Congenital) and Postnatal Hearing Loss Genetic and Chromosomal Acquired References Neurobiology 160 Overview Definition of Terms Current and Charge Voltage Resistance Capacitance Impedance Electrode/Tissue Interface Polarization Charge Transfer Charge Density Equivalent Circuits Impedance Corrosion-Stimulus Parameters Mechanisms Stimulus Parameters Scanning Electron Microscope Evaluation of Electrodes Electrical Parameters and Neural Stimulation Electrochemically Safe Stimulus Parameters Charge Density and Charge per Phase Biochemical Effects Neural Preservation Electrical Stimulation of the Cochlear Nerve Acute studies on the Effects of Low Rates of Stimulation...175

15 xviii Contents Chronic Studies on the Effects of Low Rates of Stimulation Acute Studies on the Effects of High Rates of Stimulation Chronic Studies on the Effects of High Rates of Stimulation Electrical Stimulation of the Cochlear Nucleus Acute Studies on the Effects of Low Rates of Stimulation Chronic Studies on the Effects of Low Rates of Stimulation References Electrophysiology 199 General Neurophysiology Action Potentials Strength-Duration Curves Electrical Models of the Nerve Membrane Convergence and Divergence Auditory Physiology Physics of Sound External and Middle Ear Function Cochlea Auditory Neurophysiology Electrophonic Hearing (Electrical Stimulation of the Cochlea) Mechanisms Electrophonic Hearing and Cochlear Implantation Electrical Stimulation of the Cochlear Nerve Temporal Coding Place Coding Intensity Coding References Psychophysics 296 Acoustic Stimulation Pitch and Timbre Loudness Critical Band and Ratio Musical Acoustics Bilateral Hearing Electrical Stimulation Temporal Information Temporal Information: Prelinguistically Deaf Place Information Place Information: Prelinguistically Deaf Loudness Intensity Information: Prelinguistically Deaf Musical Perception Bimodal Stimulation Bilateral Stimulation...358

16 Contents xix References Speech (Sound) Processing 381 Acoustic Articulators and Vocal Tract Shape Speech Analysis Speech Perception and Production Binaural Hearing Acoustic Models of Cochlear Implant Speech-Processing Strategies Channel Vocoders and Fixed Filters Formant Vocoders Acoustic Representation of Electrical Stimulation Speech Cues Channel Numbers Speech in Noise Channel Selection Electrical Stimulation: Principles Channel Numbers Channel Selection Speech in Noise Speech Processing Strategies Multiple-Channel Strategies: Fixed Filter Schemes Multiple-Electrode Strategies: Formant and Spectral Cue Extraction Adaptive Dynamic Range Optimization (ADRO) Dual Microphones Bimodal Speech Processing Bilateral Speech Processing References Engineering 454 Electronic and Communications Engineering Principles Speech Processors Receiver-Stimulators Bioengineering Design Principles Design Realization Conclusion References Preoperative Selection 550 Aims Adults Children...551

17 xx Contents Clinical Protocol Medical History and Examination Aims History Physical Examination Diagnosis-Etiology Adults Children Audiology Pure Tone Thresholds Impedance Audiometry Hearing Aid Evaluation Cochlear Microphonics and ABR Tests for Neuropathy Communication Speech Perception Speech Production Language Special Investigations Radiology Electrical Stimulation of the Promontory Vestibular Assessment Management Hearing and Speech Perception Predictive Factors Preoperative Counseling References Surgery 595 Overview Brief History Aims Position Multiple Electrodes Close to the Auditory Nerves Implant Electrode with Minimal Trauma to the Inner Ear Locate the Receiver-Stimulator to Allow Optimal Use of a Microphone, Speech Processor and Transmitting Coil Implant Receiver-Stimulator to be Unaffected by Growth Changes Implant Operation Performed Safely Fundamentals and Clinical Practice Preoperative Measures Incision First Stage Mastoid Cell Removal Creation of a Bed for the Receiver-Stimulator Creation of Gutter for the Lead Wire Assembly Exposure of the Round Window via a Posterior Tympanotomy...607

18 Contents xxi Cochleostomy (Opening into the Inner Ear) Insertion of Arrays Sealing the Opening Perilymph Gusher Fixing the Electrode Array and Receiver-Stimulator Flap and Wound Closure Radiology Postoperative Care Complications and Management Intraoperative Complications Postoperative Complications Special Cases Ossified Cochlea Secretory (Serous) Otitis Media Tympanic Membrane Perforation and Chronic Suppurative Otitis Media Open Mastoid Congenital or Genetic Malformation of the Cochlea Transmastoid Labyrinthectomy and Acoustic Neuroma Insertion and Reinsertion Pedestal (Plug and socket) Magnetic Resonance Imaging (MRI) References Rehabilitation 654 Aims Principles Plasticity in the Experimental Animal Plasticity Psychophysics Plasticity Cross-Modality in Humans Analytic Versus Synthetic Training Mapping and Fitting Procedures in Adults and Children Physiological and Psychophysical Principles Producing a MAP Signal Gain Loudness Summation Patient Preference Training in Adults and Children General Predictive Factors Strategy and Time Course for Learning Analytic Synthetic Environmental Sounds Background Noise...677

19 xxii Contents Music Telephone Television Mapping and Fitting Children Preprogramming Training Conditioning Initial Setting Follow-up Device Settings Neural Response Telemetry Training in Children General Personnel Pragmatics Speech Perception Perception of Environmental Sounds Speech Production Language Education of Children Acoustic Environment Strategies Program for Implanted Children Counseling of Adults and Children References Results 707 Aims Development of Tests Speech and Sound Perception: Test Principles Variability of Materials and Responses Prerecorded Versus Live Voice Training Effects and Experience Closed-Set Tests Speech Features (Consonants and Vowels) Open-Set Tests Speech Reading Speech Tracking Speech in Noise Environmental Sounds Test Batteries Questionnaires Bimodal and Bilateral Speech Processing Speech Production: Test Principles Imitative and Spontaneous Speech Computer Aided Speech and Language Assessment procedure (CASALA)...721

20 Contents xxiii Language: Test Principles Receptive Language Expressive Language Pragmatics Speech perception with Cochlear Implants Predictive Factors Speech-Processing Strategies for Postlinguistically Deaf Adults Speech-Processing Strategies for Pre- and Postlinguistically Deaf Children Speech production with Cochlear Implants Single-Channel System (3M/House) Nucleus Multiple-Channel (F0/F1/F2) and Multipeak Strategies Language Development for Pre- and Postlinguistically Deaf Children Receptive Language Expressive Language Cognition References Socioeconomics and Ethics 767 Speech and Language Benefits Biological Safety Social Benefits Personal Family School Economic Benefits Economic Measures Cost-Effectiveness Cost-Benefit Analysis Quality of Life Ethics Human Experimentation Rights of Children Attitudes of Hearing-Impaired People References Research Directions 787 Improved Sound Fidelity and Speech Processing Selection of Information Optimal Rate Stimulation Improved Coding Improved Speech Perception in Noise Bimodal Speech Processing Bilateral Speech Processing Dual Microphones Improved Speech and Language in Children Totally Implantable Cochlear Prosthesis Auditory Nerve Preservation and Regeneration References Index 813

21 Abbreviations A AAA ABF ABR AC AC ACE ADC ADRO AGC AM AN ANF ANSI AP APP ARC ASTM AVCN B BDNF BKB BM apical auditory association area adaptive beam forming auditory brainstem response alternating current auditory cortex advanced combination encoder analog-to-digital converter adaptive dynamic range optimization automatic gain control amplitude modulation auditory nerve auditory nerve fiber American National Standards Institute action potential abnormal positive potential Australian Research Council American Society for Testing Materials anteroventral cochlear nucleus basal turn brain-derived neurotrophic factor Bench-Kowal-Bamford basilar membrane BP C CA CAP CASALA CC CD CELF CF CG CI CID CIS CM CMOS CMV CN CNC CNS CRC CSF CSIRO CT bipolar capacitance compressed analog compound action potential Computer Aided Speech and Language Assessment common cavity characteristic delay clinical evaluation of language fundamentals crista fenestra common ground cochlear implants Central Institute for the Deaf continuous interleaved sampler cochlear microphonics Complementary metal oxide semiconductor cytomegalovirus cochlear nucleus consonant-nucleusconsonant central nervous system Cooperative Research Center cerebrospinal fluid Commonwealth Scientific Industrial Research Organization chorda tympani xxv

22 xxvi Abbreviations CT (scan) CUNY DAC db DC DCN DD DDE DL DLS DNA DNLL DRSP DSP EABR ECAP EcoG ED EE computed tomography City University of New York digital-to-analog converter decibels direct current dorsal cochlear nucleus data decoder data decoder/encoder difference limen development language scale deoxyribonucleic acid dorsal nuclei lateral lemniscus differential rate speech processing strategy digital signal processor evoked auditory brainstem response electrically evoked compound action potential electrocochleography electrode decoder excitatory contralateral and excitatory ipsilateral EEPROM electrically erasable programmable read-only memory EI excitatory contralateral and inhibitory ipsilateral ENT ear, nose, and throat EOWPVT Expressive One-Word Picture Vocabulary Test EPROM erasable programmable read-only memory EPSC excitatory postsynaptic current EPSP excitatory postsynaptic potential ES electrode sheath F farad F1 first formant frequency F2 second formant frequency FDA FEP FET FFT FGF FM FN GASP H HA HCRC HINT Hk HL HMM HRP Hz I IC ICC IDE IE IHC IID ILD Food and Drug Administration fluoroethylene propylene field effect transistor fast Fourier transform fibroblast growth factor frequency modulation facial nerve Glendonald Auditory Screening Procedure helicotrema hearing aid Human Communication Research Center Hearing in Noise Test hook region hearing loss hidden Markov model horseradish peroxidase hertz current inferior colliculus central nucleus of the inferior colliculus Investigational Device Exemption inhibitory contralateral and excitatory ipsilateral inner hair cell interaural intensity difference interaural loudness difference IMPEBAP implant evoked brainstem auditory potential IMSPACP Imitated Speech Pattern Test INLL INSERM IP IPSP intermediate nuclei lateral lemniscus Institut National de la Santé et de la Recherche Médicale interleaved pulse inhibitory postsynaptic potential

23 Abbreviations xxvii IPSyn ITD JFET JLD JND KEMAR khz LARSP LDL LIF LiP LL LQ LSO LVAS M MAA MAC MAIS MAP MC MDL MGB MHz MLD MLU MNTB index of productive syntax interaural time difference junction field effect transistor just discriminable level difference just noticeable difference Knowles Electronic Manikin for Acoustic Research kilohertz language assessment, remediation, and screening procedure loudness discomfort level leukemia inhibitory factor listening progress profile lateral lemniscus language quotient lateral superior olive large vestibular aqueduct syndrome middle turn minimum audible angle Minimal Auditory Capabilities (test) Meaningful Auditory Integration Scale map for the threshold and maximum comfortable levels in the speech processor maximum comfortable level minimum acceptable discomfort level medial geniculate body megahertz mesencephalicus lateralis dorsalis mean length of utterance medial nucleus of the trapezoid body MOS MOSFET MP MPP MRI mrna MSO MSP MSTP MTP MUSL metal oxide semiconductor metal oxide semiconductor field effect transistor monopolar multiple pulse per period magnetic resonance imaging messanger ribonucleic acid medial superior olive miniature speech processor Monosyllables, Spondees, Trochees, and Polysyllables Test monosyllable, trochee, polysyllable Melbourne University Sentence Lists NH&MRC National Health and Medical Research Council of Australia nhl NID NIH NINCDS NINDB NINDS NRT NST NU OC OCG OHC normal hearing level National Institute of Deafness National Institutes of Health National Institute of Neurological and Communicative Disorders and Stroke National Institute of Neurological Diseases and Blindness National Institute of Neurological Diseases and Stroke neural response telemetry Nonsense Syllable Test Northwestern University organ of Corti output current generator outer hair cell

24 xxviii Abbreviations OHUI OVE II OW P PAA PAT PB PBK PC PDGF PET PLD PLE PLS PMA PMMA PP PPLE PPVT PTA PTFE PVA PVCN QALY R RAM RC RDLS RF Ontario Health Utilities Index Orator Verbis Electris (formant synthesizer) oval window promontory polyacrylic acid Parametric Artificial Talker (formant synthesizer) phonetically balanced phonetically balanced (kindergarten) monosyllables processus cochleariformis platelet-derived growth factor positron emission tomography programmable logic device phonetic level evaluation Preschool Language Scale premarket approval Primary Measures of Music Audiation ponticulus pyramidalis phonetic and phonologic level evaluations Peabody Picture Vocabulary Test pure tone average polytetrafluoroethylene (Teflon) polyvinyl alcohol posteroventral cochlear nucleus quality-adjusted life-years resistance random access memory resistor capacitance Reynell Developmental Language Scales Radiofrequency RM RMS RNID ROC ROM RTI RW SA SAS SC SC SEM SERT SG SI SII SIT SL SM SMSP SNR SOC SP SPEAK SPL SpL SPP SPS SQUID SRT SSEP Reissner s membrane root mean square Royal National Institute of the Deaf receiver-operating curve read-only memory Research Triangle Institute round window stimulus artifact simultaneous analog system superior colliculi supporting cell scanning electron microscope Sound Effects Recognition Test spiral ganglion cells Synchronization Index Speech Intelligibility Index Speech Intelligibility Test spiral ligament scala media Spectral Maxima Sound Processor signal-to-noise ratio superior olivary complex summating potential speech processing strategy: SMSP with 20 filters sound pressure level spiral lamina single pulse per period simultaneous pulsatile stimulation superconducting quantum interference device speech reception threshold steady state evoked potential

25 Abbreviations xxix ST StV SUKL SUM SV T TA TB TESM TM TORCH UCH scala tympani stria vascularis Štátny Ústav pre Kontrolu Liečiv (State Institute for Drug Control, Slovakia) summation scala vestibuli threshold tactile aid trapezoid body transient emphasis spectral maxima tectorial membrane toxoplasmosis, rubella, cytomegalovirus, and herpes simplex University College Hospital UCSF V V VA VC VCN VCV VNLL VNTB VOT VRA XIC Z University of California at San Francisco volt vestibule vestibular aqueduct consonant-vowel ventral cochlear nucleus vowel-consonant-vowel ventral nuclei lateral lemniscus ventral nucleus of the trapezoid body voice onset time visual reinforcement audiometry commissure of the inferior colliculus impedance

26 Introduction Definition The multiple-channel cochlear implant (bionic ear) is a device that restores useful hearing in severely to profoundly deaf people when the organ of hearing situated in the inner ear has not developed or is destroyed by disease or injury. It bypasses the inner ear and provides information to the hearing centers through direct stimulation of the hearing nerve. Normal Hearing Hearing occurs when sound is transmitted down the ear canal, through the middle ear, to the inner ear. The inner ear is a very small, coiled, snail-like structure embedded in bone that houses the sense organ of hearing (organ of Corti). The organ of hearing rests on a membrane (basilar membrane) lying across the coil. This membrane vibrates selectively to different sound frequencies, so that it acts as a sound filter. High frequencies produce maximal vibrations at the beginning of the coil near an opening from the middle ear called the round window. Low frequencies produce maximal vibrations at the other end of the coil. The sense organ of hearing in the inner ear consists of cells with hairs that protrude into a gelatinous membrane. When these hairs move back and forth in response to sound, their vibrations are converted into electrical currents. This process results from chemical and physical changes in these hair cells. These electrical currents stimulate the hearing nerves and produce patterns of excitation. These patterns or stimulus codes are transmitted to the higher brain centers where they are interpreted as sound. The patterns of electrical responses are processed as pitch and loudness, as well as meaningful signals such as speech. Deafness A person who has a progressive sensorineural deafness loses the hair cells in the inner ear. As a result the hearing becomes faint and distorted and the sound has xxxi

27 xxxii Introduction to be amplified for enough cells to respond. When most of the hair cells are absent, no amount of amplification with a hearing aid will help the person hear speech, as there is no hearing organ to excite the remaining hearing nerves leading to the brain centers. At best the person will hear muffled sounds. These people are profoundly deaf and were the first who stood to benefit from the bionic ear. Overall Concept of the Bionic Ear Research that commenced at the University of Sydney in 1967 and continued at the University of Melbourne in 1970 led to a multiple-electrode cochlear implant, which was developed industrially by Cochlear Proprietary Limited in As illustrated in Figure 1, it consists of a directional microphone (a) that converts sound into electrical voltages that are sent to a small speech processor worn behind the ear (b) or a larger, more versatile one attached to a belt (c). The speech processor filters this waveform into frequency bands. The outputs of the filters are referred to a map of the patient s electric current thresholds and comfortable listening levels for the individual electrodes. A code is produced for the stimulus parameters (electrode site and current level) to represent the speech signal at each instant in time. This code, together with power, is transmitted by radio waves via a circular aerial (d) through the intact skin to the receiver-stimulator (e) implanted FIGURE 1. The cochlear implant. The components are as follows: a, microphone; b, behindthe-ear speech processor; c, body-worn speech processor; d, transmitting aerial; e, receiverstimulator; f, electrode bundle; g, inner ear (cochlea); h, auditory or cochlear nerve. (Reprinted with permission from Clark, G.M. 2000b. Sounds from silence. St. Leonards, NSW, Allen & Unwin.)

28 Introduction xxxiii in the mastoid bone. The receiver-stimulator decodes the signal and produces a pattern of electrical stimulus currents in a bundle of electrodes (f) inserted around the first turn of the inner ear (g) to stimulate the auditory nerve fibers (h). A pattern of hearing nerve activity in response to sound is produced, and provides a meaningful representation of speech and environmental sounds. The electrode bundle, lies close to, but not attached to, the spiral ganglion cells in the inner ear and their peripheral hearing nerve fibers. Training in the Use of the Bionic Ear After recovery from the cochlear implant operation, the patient attends training sessions in how to understand the sensations created by electrical simulation. The first task is to establish thresholds and maximum comfortable levels for electrical stimulation on each electrode pair. The thresholds and maximum comfortable levels are programmed into the map of the patient s speech processor. Auditory training exercises involve listening to speech and repeating what is heard. The speech material may be sentences, words, or vowels and consonants. The exercises allow the audiologist to assess the performance of the patient and at the same time provide training. The task must not be too difficult or the patient may be discouraged. The patient is also counseled on how to use the device, for example what to expect if the batteries become flat. Later, training is given in the use of the telephone. Auditory training for children concentrates on improving not only their ability to perceive and understand speech and environmental sounds, but also their speech production, receptive and expressive language, and communication. The speech material used for the training is age appropriate. The training is integrated into the child s educational program at either a preschool or school level. The children need to be taught by auditory-oral or auditory-verbal methods to take advantage of the new auditory information they are receiving. In certain situations the use of total communication where signed English is combined with an auditory stimulus will be required. Sign language for the deaf may also be used in certain children after individual assessment regarding their communication needs. Fundamental Objections and Questions In the 1960s and 1970s many believed that successful electrical stimulation of the hearing nerve to help people understand speech was not possible in the foreseeable future. A fundamental objection, which was reasonable, was that the inner ear hair cells and their nerve connections were too complex and numerous to reproduce the temporal and spatial pattern of responses in the hearing nerve by electrical stimulation with just a small number of electrodes. There are some 20,000 inner and outer hair cells required for normal hearing. A second objection was that a bionic ear would destroy the very hearing nerves

29 xxxiv Introduction in the inner ear it was intended to stimulate. For example, a Teflon strip with sharp edges can cut through the inner ear basilar membrane and lead to near-total loss of the inner ear nerve cells in the vicinity of the injury. It was also believed that the electrode could be a pathway for middle ear infection to initiate infection of the inner ear, which could in turn spread to the meningeal lining of the brain. A third objection was that speech was too complex to be presented to the nervous system by electrical stimulation for speech understanding. A fourth objection was that there would not be enough residual hearing nerves in the inner ear after they died back due to deafness to transmit essential speech information. There can be an 80% loss of the hearing nerve ganglion cells and their fibers after the destruction of inner ear hair cells in deafness. A fifth objection was that children born deaf would not develop appropriate nerve-to-brain cell connections, through lack of exposure to sound during the early critical phase of development, for electrical stimulation to give adequate hearing. The number of nerve connections on brain cells can be significantly reduced when compared to that in people with normal hearing. There were other important questions: (1) Would the electrical stimulus currents damage the hearing nerves? (2) Were the candidate materials for the implantable electrodes and receiver/stimulator toxic to tissue? (3) Would middle ear infection spread along the electrode bundle to produce infection in the inner ear with possible life-threatening infection around the linings of the brain (meningitis)? (4) Could electrodes be inserted into the inner ear far enough so that the hearing nerves responsible for the place coding of speech frequencies would be stimulated? (5) What type of patients should be selected? (6) How should the operation be performed? (7) Would the perception of pitch on a multiple-electrode or place-coding basis be possible? (8) Would the perception of pitch on a timecoding basis be possible? (9) What electrical currents would produce loudness? (10) Would patients have memory for sounds and speech after prolonged deafness? (11) Could speech be processed so that patients could understand conversations? (12) Would speech and music sound natural? (13) If a speech-processing scheme was achieved for English, would it be effective in other languages? (14) How important a factor was the child s age at implantation with regard to learning to understand speech? Answers to the Fundamental Objections The first fundamental objection, that the inner ear hair cells and their nerve connections were too complex and numerous to reproduce the temporal and spatial pattern of responses in the hearing nerve by electrical stimulation with just a small number of electrodes, was studied by determining how well electrical stimulation could reproduce the coding of sound. The temporal coding of frequency was examined in the experimental animal by determining how well groups of brain cells could respond at increasing rates of stimulation. The voltages from brain cells and brainstem field potentials at increasing rates of stimulation showed the

30 Introduction xxxv electrical activity in the auditory brainstem was markedly suppressed by stimulus rates at 100 pulses/second. Behavioral studies in the experimental animal showed that rates of stimulation in excess of 200 to 600 pulses/second could not be discriminated. The experimental animal findings thus indicated that the reproduction of the temporal coding of frequency by electrical stimulation with a single-electrode cochlear implant could reproduce speech frequencies only from 200 to 600 cycles/ second, which is much less than the 4000 cycles/second needed for speech intelligibility. Therefore, the best chance of helping deaf people understand speech was to use multiple-electrode stimulation to provided more information for speech understanding. To achieve the place coding of frequency through multiple-electrode stimulation required determining where to place the electrodes in the inner ear so that the current would most easily pass through separate groups of hearing nerve fibers connected to the different frequency regions of the brain. Research showed that the compartment below the sense organ of hearing (scala tympani) and close to the ganglion cells at the center of the inner ear spiral was the correct location. Research also demonstrated that electrical currents could be partly localized to groups of nerve fibers within the inner ear without it short-circuiting away through fluid by pushing electrical current out one electrode and pulling it back from another (bipolar stimulation). The animal experiments referred to above demonstrated that both temporal and place frequency coding or pitch perception could be only partially reproduced by electrical stimulation. In other words, a cochlear implant is like a bottleneck between the world of sound and the central hearing pathways of the brain. The second fundamental objection was that if an electrode was implanted in the inner ear, which was particularly important for multiple-electrode stimulation, it would damage the very nerves it was intended to stimulate. It was found, however, in the experimental animal, that if no excessive force was used with its insertion, the hearing nerves were preserved. The risk of injury was reduced to a minimum if the electrode bundle had the right mechanical properties. It needed to be smooth, tapered, flexible at the tip, and stiffer toward the proximal end. Infection could be restricted from entering the inner ear if the electrode entry point was sealed with a graft of fascia, and care was taken to prevent infection of the middle ear during the healing phase of the tissue over the first few weeks postoperatively. The density of the electrical charge passing through electrodes with electrical stimulation was also known to damage nerve fibers. The safe limits for use with a cochlear implant had to be tested, too. It was found to be safe if the current had a positive and negative phase to reduce the buildup of direct current (DC), and the charge density was below approximately 32 microcoulombs per square centimeter per phase. The third objection, that speech was too complex to be presented to the nervous system by electrical stimulation for speech understanding, would have to be addressed by multiple-electrode stimulation to transmit as much information as pos-

31 xxxvi Introduction sible through the bottleneck. This required studies on patients to determine how effective multiple-electrode stimulation would be, as speech perception is an especially human skill and could not be evaluated on the experimental animal. Studies on patients required developing a fully implantable receiver-stimulator to receive information transmitted through the intact skin, rather than a plug and socket, which was more likely to break and become infected. A prototype receiver-stimulator to use on patients was produced by the University of Melbourne from 1974 to 1978 using hybrid technology that connected a number of silicon chips together on a silicon substrate or wafer. The wafers were placed in a watertight or hermetically sealed container. The prototype receiver-stimulator was implanted in the first profoundly deaf adult patient on August 1, 1978, with the banded electrode array passing around the inner ear to lie near, but not in direct contact with, the nerves relaying speech frequency to the brain. Perceptual studies were then undertaken on the first and subsequent patients to determine if the findings on the temporal and place coding of frequency in the experimental animal were applicable to humans. The patient studies confirmed that rate of stimulation was not effective in transmitting frequency or pitch information over the range required for speech understanding. Pitch ratios were plotted against repetition rate, and it was shown that when the pitch of a stimulus was compared with a reference rate of 100 pulses/s, the pitch ratios increase linearly up to 300 pulses/s, and then reached a plateau; 300 pulses/s is much less than the 4000 pulses/s needed for speech understanding. The studies on the place coding of frequency showed that with localized electrical stimulation the patients could perceive only timbre, not true pitch. In the high-frequency areas of the inner ear the sensation was sharp, and on the lower frequency side it was dull. However, the patients could rank the timbre according to the site of stimulation. The perceptual studies on the patients confirmed the findings on experimental animals that electrical stimulation with the cochlear implant was a bottleneck for information from the outside world to the central auditory pathways. The first research to transmit information through the bottleneck selected speech frequencies using fixed filters with similar properties to the tuning of the inner ear. When the outputs were used to stimulate the hearing nerves simultaneously, the result was poor. Simultaneous stimulation produced overlap of currents resulting in unpredictable variations in loudness. However, a speech-processing strategy was discovered that gave the patients the ability to understand connected or running speech when presented with speech reading or even using electrical stimulation alone. The clue to this speech-processing strategy came when the first patient reported vowel sounds when each electrode was stimulated on a place-coding basis. The vowels corresponded to those perceived by normal-hearing people when similar areas of the inner ear were excited by single-formant frequencies. Formants are concentrations of energy at particular frequencies or vocal tract resonances. They are important for intelligibility, especially the second formant. This research led to the University of Melbourne s inaugural speech-processing