|Year : 2003 | Volume
| Issue : 4 | Page : 371-378
|Imaging of cochlear implants
IK Indrajit, JD Souza, VK Singh, E James, S Badhwar
Classified Specialist (Radiodiagnosis), Mumbai-400005, India
Click here for correspondence address and email
| Abstract|| |
Traditionally, cochlear implants are restrictedly perceived by radiologists as devices absolutely contraindicated for MRI studies. With increasing cochlear implantations carried out world over, and the increasing availability of Digital Radiography and Multislice CT, imaging plays a key role, both in preoperative period and following implantation . Essentially, cochlear implants are medical devices that electrically stimulate the auditory nerve in the cochlea. Designed to allow patients with severe hearing loss to perceive sound, cochlear implant systems contain internal electrode array that is placed inside the cochlea. Computed tomography (CT) is considered the modality of choice for accurate imaging of the bony labyrinth. Of late, Multislice CT applications not only accurately displays complex three-dimensional anatomic structures of inner ear and contents of middle ear cavity, but also generates interactively volume rendered images of labyrinth. This review article demystifies Cochlear Implants and illustratively describes the role of imaging generated by Digital Radiography and Multislice CT.
Keywords: Cochlear Implants; Multislice CT; Volume Rendering
|How to cite this article:|
Indrajit I K, Souza J D, Singh V K, James E, Badhwar S. Imaging of cochlear implants. Indian J Radiol Imaging 2003;13:371-8
|How to cite this URL:|
Indrajit I K, Souza J D, Singh V K, James E, Badhwar S. Imaging of cochlear implants. Indian J Radiol Imaging [serial online] 2003 [cited 2019 Sep 21];13:371-8. Available from: http://www.ijri.org/text.asp?2003/13/4/371/28710
| Introduction|| |
The cochlea is a coiled structure of 23/4 turns containing three parallel fluid channels : outer scala vestibuli, inner scala tympani, and a central cochlear duct (scala media). A cochlear implant is a small, complex medical electronic device that improves speech recognition to a profoundly deaf patient, who derives little benefit from hearing aids . It is designed to stimulate the spiral ganglion cells in patients by insertion of an electrode array into the scala tympani or scala vestibuli. Referred popularly as Bionic Ear, it comprises of a microphone, that picks sound from environment, a speech processor that arranges sounds picked up by microphone, a transmitter and receiver/ stimulator that receive signals from speech processor converting them into electric impulses and electrodes that collect impulses from stimulator and send them to brain. An implant does not restore or create normal hearing; rather it gives a deaf person an auditory understanding of environment and assisting in the understanding of speech.
| Components of cochlear implants|| |
The evolution of Cochlear Implants has resulted in two types of implants - single and multiple channels [Table - 1]. Multichannel cochlear implants give better speech understanding than single-channel devices. They are intended to make use of tonotopic organization of cochlea by selectively stimulating subpopulations of the auditory nerve. The two earliest models of multichannel intracochlear implants approved by FDA for adults and children in the United States were Nucleus 22 (Cochlear Corp) and the Clarion (Advanced Bionics Corp). At our institution Nucleus multichannel cochlear implants have been used. [Table - 2].
The widely used multichannel cochlear implant prosthesis has two parts: an externally worn component and a surgically implanted component. The externally worn component comprises of an ear-level microphone, a speech processor carried in a pocket, and a transmitter placed behind the ear while the implanted components consists of a receiver/stimulator and an electrode array [Figure - 1]A. The receiver/stimulator is placed within a drilled calvarial well beneath postauricular soft tissues, while the electrode array is selectively inserted into the scala tympani through a cochleostomy. All external components are connected to one another by electrical wires, while the transmitter and receiver are usually connected by a pair of magnetic disks. During implantation technique, the electrode array is introduced by a transmastoid facial recess, through a cochleotomy anteroinferior to the round window, into the scala tympani of the basal turn
| Mechanism of Cochlear implants|| |
Put in a nutshell, the pathway of sound energy information commences at the microphone, and is then serially transmitted to external speech processor, transmitting coil, the implant, the 22 electrodes, the nerve endings in cochlea and finally to the brain along the auditory nerve. The 22 electrodes replace the function of many thousand hair cells in a normally hearing ear.
Thus a microphone detects sounds and speech. Speech processors filters, analyzes and digitizes the received speech and environment sound into coded signals, which are subsequently sent to a transmitting coil. The coil sends these signals as FM signals to the cochlear implant under the skin. Thereafter the cochlear implant delivers appropriate electrical energy to the electrode array inside the cochlea, which stimulate the nerve fibers within and electrical sound information is sent through auditory system to brain for interpretation. The position of stimulating electrodes within cochlea will determine the frequency or pitch of sounds. The amount of electrical current will determine loudness of sounds. Different parameters of speech and environment sound such as different pitch, loudness and timing cues can be adjusted. The speech processor is not a hearing aid which amplifies sounds, rather it selects information in speech signal and generates a pattern of electrical pulses as close as possible to original speech sound that is sent to brain for interpretation. Thus cochlear implants compensate for damaged or non-functioning parts of inner ear.
| Indications and Contraindications|| |
The indication for Cochlear Implant in adults include bilateral severe-to-profound sensorineural hearing loss with 70 dB hearing loss or greater with little benefit from hearing aids for 6 months, psychologically suitable, with no anatomic contraindications and no medical contraindication. In children, the indications for Cochlear Implantation includes bilateral severe-to-profound sensorineural hearing loss with PTA of 90 dB or greater in better ear, no appreciable benefit with hearing aids, age greater than 12 months, no medical or anatomic contraindications and presence of motivated parents.
Contraindications for cochlear implant comprise of incomplete hearing loss, mental retardation, psychosis, organic brain dysfunction, active middle ear disease, CT findings of cochlear agenesis (Michel deformity) or small Internal auditory canal, mastoidectomy, labyrinthitis ossificans and advanced otosclerosis. Importantly dysplasia is not a contraindication. A list of contraindications demonstrable by CT is listed in [Table - 3]
| Protocol and Methods|| |
Five children (three male, two females; mean age 5 years) were examined with Multislice CT for pre and post cochlear implant evaluation, after careful screening and meticulously selection by definitive clinical criteria. The parents were motivated and counseled about the possible results, which ranged from perceiving only sound to understanding speech.
Digital Radiography was performed using a Siemens Polystar DSA1000 MA machine at Department of Radiodiagnosis & Imaging. Views were routinely taken in AP, Stenver, Modified Chausse III.
The CT examinations were performed with a Multislice CT scanner (Volume zoom; Siemens Medical Systems, Erlangen, Germany) at our institution. The patient was positioned in the center of the scan field of view in each case. A spiral mode for inner ear using 2 x 0.5 mm slice collimation and a slice width 0.5 mm was applied. The axial images were reconstructed with an ultra highresolution kernel algorithm. Reconstruction was performed with steps of 0.3 mm because the cochlea measures only about 5mm in diameter. An FOV of 150 mm was used to allow comparison of both sides. The average scan time was 42 s. We used the same protocol to evaluate the patient before and after implant insertion. The best } images are obtained by use of high-resolution software algorithm with smallest possible section thickness, reconstruction of overlapping sections, and a narrow FOV mainly focusing on target structures.
| Postprocessing|| |
After transferring the image data to a high-end graphics workstation (Siemens Leonardo), 3D visualization based on basic and advanced interactive rendering was performed. Multiplanar Reformation, Curved Multiplanar Reformation, Maximum intensity projection, Shaded Surface Display and Volume rendering was applied to visualization of the structures of the temporal bone in each case. In all cases, clear representations of the bony labyrinth and the facial canal were provided. Using volumerendering capabilities at high-end graphics workstation, semitransparent representations of structures of inner ear were obtained. While no segmentation was performed before rendering and prior to visualization process, color and opacity values were adjusted interactively to delineate all structures related to the inner ear in real-time. After optimal delineation of target structures the color and opacity table were saved for application in future studies. Importantly, the post processing 3d imaging procedure, commencing from transfer of data, rendering of images, labeling and filming was performed in 30 minutes, promoted by tremendous acceleration by hardware and user friendly interface.
| Evaluation|| |
A pre-implantation CT check list was used to systematically review the temporal bone, as depicted in [Table - 4]. Close attention was given to status of round window, cochlear turns, facial nerve. Additionally, any potential contraindications such as obstruction of labyrinth by ectopic bone or fibrous tissue due to meningitis or labyrinthitis, otosclerosis, mastoiditis, were closely looked for. CT scan in post implant cases was evaluated for cochlear lengths and array insertion depth both of which assumes importance in the final outcome of word recognition by implant recipients. The array of multichannel devices are inserted 22 to 26 mm into the middle cochlear turn.
| Discussion|| |
Imaging plays an important role in pre-operative assessment and post-operative review of cochlear implants candidates . With a shift in the procedure of cochlear implants from tertiary referral centers to community practices, it is an established fact that radiologists are increasingly called upon to play a key role in preoperative selection of candidates as well as assess post cochlear implantation patients. At the moment, the modalities that are available includes Radiography, Digital Radiography, CT , while MRI is restricted to pre implantation assessment.
Customarily in most cases, plain radiography is the first modality used for evaluation. The radiography of cochlea or cochleography can assess the position of the implant and insertion of the stimulating electrode into the cochlea. The views include AP projection, Stenver's, and modified Chausse III projection. Most views require grid, kVp range of 70-80, small focal spot, collimation
In Stenver's view, the patient is in prone position with IOM line perpendicular to the table. From a true lateral position rotate head 45 degrees toward side of interest, with the forehead, nose, & zygoma resting on table. The ray should exit about 1 inch anterior to the external auditory meatus placed on the table. The central ray is angled 12 degrees cranially. An ideal Stenver's view image will show the ipsilateral petromastoid (closest to cassette) in profile without distortion. The structures that form the boundaries will be lateral border of skull to lateral border or orbit. The mastoid process is seen in profile below cranium. The ramus of the mandible will superimpose cervical spine, while condyle of the mandibular will be near the petrous bone
In recent times digital radiographs have been increasingly used in view of their ability to delineate the cochlear anatomy better. Specialised views such as a modified Chausse III projection performed exclusively in digital techniques can even determine the degree of electrodes rotation of multichannel cochlear implants within the cochlea  as depicted in [Figure - 1]A & B. The modified Chausse III projection is obtained with the skull placed supine on radiography table with infraorbitomeatal line perpendicular to film cassette. After rotating the skull 30 degrees away from the side to be examined, the central X-ray beam is angled 15 degrees cephalad to the infraorbitomeatal line. On these radiographs, an electrode is considered as inserted completely if all electrode contacts were projected medial to a line drawn through the superior semicircular canal and the vestibule. A normal placement of implant is indicated by demonstrating the electrodes to be regularly spaced and curving gently within the first turn of the cochlea. From radiography, the insertion depth is determined by counting the number of electrodes that projects medial to the cochlear promontory.
Postoperative digital radiography demonstrates not only electrode position but confirms satisfactory intra-cochlear electrode placement . Moreover, the number of active inserted electrodes can be determined, while complications such as electrode kinking and slippage can be identified. The digital technique is comfortable for patient, easily reproducible, offers high quality images, localises electrodes, all at lower radiation dose in comparison to conventional radiography or CT scan.
The role of CT imaging is not confined merely to the preoperative assessment of the inner ear while ruling out conditions that are potential contraindications, but also extends beyond to the evaluation of the postoperative cochlea. High resolution CT with thin (0.5 mm to 1 mm) sections in axial and coronal planes of petrous temporal bones is commonly used in imaging pre and post cochlear implant candidates . Multislice CT offers additional advantages of isotropic imaging with facilities for interactive volume rendering. In general, information about threedimensional (3D) anatomic structures is offered as twodimensional (2D) images.
A pre-implant CT imaging has few important roles. Firstly, it demonstrate conditions, which preclude cochlear implant procedure [Table - 2]. Some of the common contraindications are severe otosclerosis and postmeningitic labyrinthitis, obliterative labyrinthine ossification, congenital cochlear malformation, temporal bone fractures and severe cochlear or fenestral otosclerosis ,. Secondly, the most suitable side for operation can be selected for cochlear implants candidates. Thirdly, CT imaging identifies possible barriers to the mechanical insertion of the internal components of cochlear implants as well as depicts normal variants. This feature is further enhanced with Multislice CT where the delineation of anatomic structures for surgical planning is much improved [Figure - 2].
In precochlear implant evaluation, it is prudent to employ a systematic evaluation of the temporal bone, as depicted in [Table - 2]. Few key objective anatomical parameters that need carefully evaluation includes the patency of basal turn of the cochlea, the shape, size and orientation of the round window niche, the course of the facial nerve, the proximity of carotid artery and jugular vein, the parietal bone thickness for housing the receiving device and the degree of pneumatization of the mastoid. CT scan also offers information on the size of the vestibular aqueduct, cochlear aqueduct, and internal auditory canal. Besides, details of the carotid canal which lies in close topographic relationship to the basal turn of the cochlea is also clearly depicted. The bony wall between them can be as thin as 0.2 mm or as thick as more than 6 mm.
It has been suggested that one of the many factors that is causing variability in word recognition amongst implant recipients is the position of the electrode array in the cochlea. A study revealed that word scores significantly correlated with insertion depth as a percentage of total cochlear length. To this end, the objective analysis of cochlear lengths and array insertion depth assumes -' nportant .
| Multislice CT Technology and Rendering Techniques|| |
The advent of Multislice CT technology has introduced benefits such as improved temporal resolution, improved spatial resolution in z-axis, decreased image noise, efficient x-ray tube use, and longer anatomic coverage. Taken together, these factors have largely improved the diagnostic accuracy of CT applications.
In temporal bone imaging, Multislice CT has brought in overwhelming changes. Firstly, multislice CT allows for faster acquisition times (up to eight-fold) than conventional single-slice spiral CT scanners, a feature permitting larger anatomic coverage with thin-slice collimation protocols. This enables greatly enhanced spatial resolution along the longitudinal axis, providing isotropic 3D voxels and generation of volumetric data sets. Consequently, a greater portion of z-axis is covered in a shorter amount of time, while ultra-thin slices can be obtained (1 mm), that can be reconstructed at small intervals (1 mm), producing high-resolution 3D images of the entire inner ear.
The second key area relates to ongoing software and hardware refinements in 3D reconstruction. A combination of dedicated workstations, advanced rendering techniques, interactive evaluation and "feature-rich" software, makes 3D reconstruction easier and faster. By rotating the data set in any plane on the workstation, the entire inner ear or its components, the facial nerve, the ossicles, the round and oval windows are displayed and evaluated as one image.
Before the advent of Multislice CT and advances in rendering techniques, temporal bone imaging was confined to limited evaluation of axial slices. The complex anatomy of temporal bone, including inner ear, could not be evaluated in its true perspective. However, rapid advances in technology have changed this, generating 03D images of temporal bone, inner ear, bony and membranous labyrinth in living subjects available routinely.
3D CT representations of the inner ear often generate spectacular images of the labyrinth, the ossicles and the implant. This is made possible by a combination of highend graphics workstations whose software accelerated by hardware allows real-time manipulation of target structures within the image data set using a combination of tools such as clip planes, segmentation extraction, cross-referencing tool, flexible reconstruction of geometric models, interactive measurement of point distances within 3D. A standard CT combined with interactive volume rendering enables fast and flexible display of 3D images.
Available to us at the moment are few novel postprocessing techniques [Table - 5]: multiplanar reconstruction (MPR), maximum intensity projection (MIP), shaded-surface display (SSD), and volume rendering (VR).
Multiplanar Reformation is the simplest 3D technique commonly employed as an adjunct to viewing axial images. The reformation enables creation of single section 2D and thick slab 3D images [Figure - 3]. The axial sections along with volume of interest are stacked to generate an imaging volume. To this imaging volume, a reformatting algorithm is applied and arbitrarily rotated. Multiplanar Reformation has evolved bringing in a new feature called "Curved oblique Multiplanar Reformation". Here a defined curved line in a single-voxel-thick plane, orthogonal to a reference plane, is extruded by computer algorithm through the entire data set. The resultant "curved plane" is flattened and displayed as a 2D, composite image representing the target vessel.
Shaded Surface Display offers opaque surface representations of the labyrinth. User-selected upper and lower thresholds are defined to specific range of Hounsfeld units that need to be displayed. Light effects with the presence of shadows give depth to the image. Studies have shown that owing to partial volume effects, the bone structures surrounding the inner ear are difficult to eliminate without extensive editing.
Maximum Intensity Projection displays only brightest voxel along every ray of sight, with darker voxels in front or behind not displayed. This leads to loss of depth information with preservation of some of density information. This type of rendering does not require selection of any thresholds. Two problems associated with MIP includes a) displayed voxels are not necessarily part of the labyrinth or target structures and b) parts of a structure may disappear with only slight changes in viewing direction. These can create errors in interpretation. In recent times MIP as a rendering method has limited value in CT in comparison to MRI, due to the distribution of Hounsfeld data values within the hypoattenuating structures of temporal bone.
Volume Rendering uses all attenuation data information contained inside a volume of tissue. The technique, which is dependant on fast computer processing, assigns a specific color and opacity value to every attenuation value of CT data. The term virtual labyrinthoscopy has been suggested for visualization of the labyrinth by using volume rendering. A high opacity produces low transparency while low opacity produces high transparency. The temporal bone and its labyrinthine structures can be displayed as nontransparent images similar to SSD images or semitransparent 3D representations. Creditably, within the semitransparent representation, the transition between soft tissue and bone structures is clearly displayed thus taking advantage of the partial volume effect. In contrast to SSD and MIP it allows generation of more meaningful images. [Figure - 4]A and B
| Common Pitfalls in Imaging|| |
While imaging has an important role in the management of cochlear implants, it is not without its shortcomings. Radiographic techniques fail to demonstrate fibrous or bony obstruction of the basal turn of the cochlea, which can cause difficulties during cochlear implantation. Fibrotic or osseous obstruction of one or both scala channels, is a common sequela to meningitis or labyrinthitis which choke the fluid-filled channels necessary for implant insertions .
Similarly, CT evaluation is occasionally limited by its failure to identify blockage of cochlear canal due to fibrosis, bought about by lack of demonstrable density changes. Such cases can be potentially submitted to MRI. One study showed that nearly 69% of patients were found to have some degree of cochlear ossification with normal CT and recommended that otologists should expect bony obstruction within basal turn of cochlea even if CT scan is normal .
The newer 3D visualization techniques has few problems too. The bony labyrinth is difficult to visualise due to the tiny size of the structures and their location within the temporal bone, which consists of many air-filled cells. In view of the contiguously located, extreme values of Hounsfeld unit (dense bone adjacent to air), it is almost mandatory to visualize the structures of the inner ear with time-consuming and extensive segmentation . A major problem in 3D visualization is delineation of structures of the inner ear with objects of similar attenuation in vicinity, making differentiate the target structures difficult. This problem is solved to an extent by selecting applying clip planes, which suppress disturbing structures within the 3D representation. and rendering small volumes of interest.
Finally, the issue of radiation exposure risks assumes significance particularly in evaluation of children. A study comparing different modalities used for confirmation of electrode position, showed that the radiation exposure by digital radiography, conventional radiography and CT scanning was 470 microGy, 40 to 440 microGy and 950 microGy, respectively .
| Role of MRI|| |
Like CT, MRI is helpful in the diagnosis of congenital and acquired changes within the inner ear. However, the experience in routine preoperative MR imaging is still limited. It must always be remembered that MRI has a role restricted to pre-implant evaluation and is an absolute contraindication following implant insertion. Heavily T2weighted images, obtained either by a fast spin-echo or a constructive interference steady state technique, enable detection of cochlear fibrosis, usually not visible on CT scans . Besides acute inflammation is also visible on T2-weighted images. MRI also displays the modiolus and the size and course of the cochlear nerve, besides delineating the facial nerve. Congenital defects in the internal auditory canal, the cochlea and the vestibule can be diagnosed and also quantified.
At the moment, MR imaging has a limited role, not entirely replacing CT in preimplantation examination. This situation is likely to stay in view of a) the increased cost element to an already expensive procedure b) the need for sedation and monitoring of children and c) poor delineation of bony mastoid bed.
| Conclusion|| |
Cochlear implants are sophisticated medical devices1 inserted operatively in severe-to-profound sensorineural hearing loss. Imaging plays a key role, both in preoperative period and following cochlear implantation. The role of imaging has further increased by a combined use of conventional radiography, digital radiography, multislice CT. MRI is presently restricted strictly to evaluation of preimplant candidates. CT is considered the modality of choice for accurate imaging of the bony labyrinth. Recently, Multislice CT applications have ushered in accurately display of 3D anatomic structures of inner ear with generation of interactively rendered images of labyrinth. Virtual labyrinthoscopy is the new approach to interactive 3D analysis of the inner ear . However imaging by CT is limited by its inability to identify fibrosis and obliteration of cochlea. It is expected that in future, versatile 3D representations of the labyrinth will be available in CT and MRI, which will push further the frontiers of labyrinthine and implant imaging.
| References|| |
|1.||Phelps PD, Annis JA, Robinson PJ. Imaging for cochlear implants.Br J Radiol 1990 63(751):512-6 |
|2.||Rosenberg RA, Cohen NL, Reede DL. Radiographic imaging for the cochlear implant Ann Otol Rhinol Laryngol 1987; 96(3 Pt 1):300-4 |
|3.||Shpizner BA, Holliday RA, Roland JT et al. Postoperative imaging of the multichannel cochlear implant. Am J Neuroradiol 1995;16(7):1517-24 |
|4.||Lawson JT, Cranley K, Toner JG. Digital imaging: a valuable technique for the postoperative assessment of cochlear implantation. Eur Radiol 1998;8(6):951-4 |
|5.||Mueller DP, Dolan KD, Gantz BJ . Temporal bone computed tomography in the preoperative evaluation for cochlear implantation. Ann Otol Rhinol Laryngol 1989;98(5 Pt 1):346-9 |
|6.||Woolford TJ, Roberts GR, Hartley C, Ramsden RT. Etiology of hearing loss and cochlear computed tomography: findings in preimplant assessment. Ann Otol Rhinol Laryngol Suppl 1995 ;166:201-6 [PUBMED] |
|7.||Skinner MW, Ketten DR, Vannier MW, Gates GA, Yoffie RL, Kalender WA. Determination of the position of nucleus cochlear implant electrodes in the inner ear. Am J Otol 1994;15(5):644-51 |
|8.||Balkany TJ, Dreisbach JN, Seibert CE. Radiographic imaging of the cochlear implant candidate: preliminary results. Otolaryngol Head Neck Surg 1986;95(5):592-7 |
|9.||Seidman DA, Chute PM, Parisier S. Temporal bone imaging for cochlear implantation. Laryngoscope 1994; 104 (5 Pt 1):562-5 |
|10.||Tomandl BF, Hastreiter P, Knut E. W et al Virtual Labyrinthoscopy: Visualization of the Inner Ear with Interactive Direct Volume Rendering Radiographics. 2000;20:547-558 |
|11.||Hansa P, Granta AJ, Laitta RD et al Comparison of ThreeDimensional Visualization Techniques for Depicting the Scala Vestibuli and Scala Tympani of the Cochlea by Using High-Resolution MR Imaging Am J Neuroradiol 20:1197-1206 |
I K Indrajit
Classified Specialist (Radiodiagnosis), Mumbai-400005
Source of Support: None, Conflict of Interest: None
[Figure - 1], [Figure - 2], [Figure - 3], [Figure - 4]
[Table - 1], [Table - 2], [Table - 3], [Table - 4], [Table - 5]
|This article has been cited by|
||An analysis of medical image processing methods for segmentation of the inner ear
| ||Todd, C., Kirillov, M., Tarabichi, M., Naghdy, F., Naghdy, G. |
| ||Proceedings of the IADIS International Conference Computer Graphics, Visualization, Computer Vision and Image Processing. 2009; : 213-218 |
| Article Access Statistics|
| Viewed||7110 |
| Printed||261 |
| Emailed||11 |
| PDF Downloaded||0 |
| Comments ||[Add] |
| Cited by others ||1 |