ABSTRACT INTRODUCTION

Size: px

Start display at page:

Download "ABSTRACT INTRODUCTION"

Jonathan Warner
5 years ago
Views:

1 Three-dimensiol reconstruction and finite element modeling for electrical stimulation of human brainstem J. Huang & J.P. Mobley Department of Engineering, House Ear Institute, 20057, ABSTRACT Topographical and three-dimensiol models of the brainstem segment including the cochlear nuclei were constructed employing computer aided design software. The geometric information was then utilized to construct a finite element model to describe the electric potential distribution in the area for monopolar stimulation. The model demonstrates the effects of ground electrode position on the electric field geometry. Placing the electrode further medial inside the lateral recess increased stimulation of the cochlear nuclear complex. INTRODUCTION Finite element methodology was first introduced twenty years ago in structural mechanics. Since then it has been applied to solve problems in many areas where system behavior is described by differential equations. The field of biomechanics has benefited tremendously from this technique, (Spilker and Simon, 1988). In recent years, finite element techniques have been applied to bioelectric.phenome including direct electrical stimulation of human tissues, (Miller and Henriquez, 1990; Finley et al., 1990; Sepulveda et al., 1983). This paper describes the electric field in the brainstem tissue stimulated by a monopolar central electroauditory prosthesis. The auditory brainstem implant (AJBI) is designed to provide auditory sensations for patients with severed auditory nerves. During acoustic neuroma surgery the cochlea and auditory nerve is removed to gain access to the acoustic tumor mass. A special electrode is inserted into the lateral recess at the time of tumor removal for cochlear nuclear complex (CNC) stimulation. The complex includes the dorsal cochlear nucleus (DCN) and the ventral cochlear nucleus (VCN), which relay the auditory sigls to the cerebral cortex. Speech sigls are collected by the microphone, processed by the sigl processor, and fed into the electrodes through a percutaneous plug or transcutaneous coil. Because of the complexity of the brainstem structures and the iccessibility of the stimulating area for measurements and evaluation, three dimensiol reconstruction of the brainstem and finite element alysis are used to describe the electric fields.

2 184 Computatiol Biomedicine MATERIALS AND METHODS Two dimensiol histologic slides of the human brainstem showing the CNC and surrounding nuclei were prepared by the neuroatomy lab (Figure 1). The graphic model was then converted into IGES file format and brought into a finite element alysis software AFEMS running in a PC environment (FEM Engineering, Inc.). Ventral Cochlear Nucleus Dorsal Cochlear Nucleus Vestibular Nerve Spinothalamic and Spil Trigemil Tract: Lateral Vestibular Nucleus Medial Vestibular Nucleus Facial Nucleus Superior Olive Trapezoid Body Figure 1. A section of histologic slide. complex. Shaded area is the cochlear nuclear Projection drawings were made of seven slides which included the cochlear nuclei. The spacing between each histological slide was inch. The method of reconstruction of a 3D image of the brainstem was summarized in a previous presentation. (Huang, Mobley, and Moore, 1993). The data of the brainstem outline were saved in IGES format. The IGES files were then loaded into the AFEMS's solid modeler. The 3D model was constructed by connecting the brainstem outline of each layer into a surface and extruding it in the Z axis for ". The solids were numbered in descending order. 3D meshes were generated on the seven solids in the units of 15 x 10 x 1 along the dorsal-ventral, medial-lateral, and rostral-caudal directions respectively. There were 1050 elements and 2464 nodes in the model. Since the solids had contacting faces, the coincident nodes were combined by the program.

3 Computatiol Biomedicine 185 Trapezoid Body Superior Olive VIII Nerve Ventral Cochlear Nucleus Lateral Recess/ Dorsal Cochlear Nucleus Figure 2. Views of wireframe reconstructed brainstem with auditory structures and the area of lateral recess. Ventral Cochlear Nucleus Dorsal Cochlear Nucleus Lateral Vestibular Nucleus Cochlear Nerve Medial Vestibular Nucleus Vestibular Nerve Facial Nucleus Spi no t.h aim iii c and Spil Trigemil Tracts Trapezoid Body Inferior Olive Figure 3. A view of the 3D reconstructed brainstem.

186 Computatiol Biomedicine The equations used for the electric-field distribution are detailed in the work by Sepulveda et al., 1983. Tissue conductivity was set at 0.03628 per ohm.

4 186 Computatiol Biomedicine The equations used for the electric-field distribution are detailed in the work by Sepulveda et al., Tissue conductivity was set at per ohm.cm, based on the paper by Geddes and Baker (1967). The maximum neural stimulating amplitude used by the ABI is not more than 20 microcoulomb/cnf/phase (Eisenberg et al., 1986; McCreery et al., 1990; Shannon, 1992). In the model a DC current source with charge density of either 20 microcoulomb/cnf or 10 microcoulomb/cnf was applied to a group of elements adjacent to the lateral recess. A remote reference was set to zero. Shifting of electrode band position medially was tested in the model. Isopotential contours were drawn to distinguish stimulating areas and sigl levels were compared to data in the neural stimulation literature. RESULTS A wireframe image of the reconstructed brainstem in four viewing angles is shown in Figure 2 with only the auditory structures displayed. The rectangle represents the area of lateral recess where the surface electrode is placed. Figure 3 shows the reconstructed brainstem with other functiol nuclei. Side effects experienced by the patients are related to stimulation of these other nuclei. (Brackmann et al., 1993). For example, stimulation of the cranial nerve Vn and IX can cause 'unpleasant facial twitches or tickling sensation at the throat. Stimulation of the spinothalamic and spil trigemil tracts (STT) is responsible for pain and temperature sensations. The vestibular structures, including the medial vestibular nucleus (MVN) and lateral vestibular nucleus (LVN) can result in dizziness and usea when stimulated. Figure 4a shows the potential distribution when the reference was the far ventral-medial element in Area #1. The active electrode was at the lateral end of the lateral recess (Area ), and the source charge density was 10 microcoulomb/cnf. The reference in Figure 4b is the ventral end element of the third row medial to the midline (Area #1). When overlaying the histologic sections (See Figure 1) with the corresponding extruded solids on which potential contours are plotted, we can tabulate the relation of propagating field and the nuclei location as in Table 1. CNC consistently receives higher sigls than other non-auditory nuclei. To examine the level of the stimuli, the charge density is doubled to 20 microcoulomb/cnf. The fields take similar shapes as in Figure 4ab, but the potential levels are higher for each area. When the stimulating electrode is shifted medially inside the lateral recess, the field moves accordingly, but the potential levels outside the CNC area are smaller (Figure 4c). This explains why fewer side effect occur in patients when the electrode is placed more medially (Brackmann et al., 1993).

POTENTIALS (MICROVOLT) 10 > 8 > 7 > 6 > 5 > 4 > 3 > 2 > 1 > +1.69E+00 +1.51E+00 +1.32E+00 +1.13E+00 +9.48E-01 +7.62E-01 +5.76E-01 +3.89E-01 +2.03E-01 +1.

5 POTENTIALS (MICROVOLT) 10 > 8 > 7 > 6 > 5 > 4 > 3 > 2 > 1 > +1.69E E E E E E E E E E-02 Computatiol Biomedicine 187 POTENTIALS (MICROVOLT) 10 > 9 > 8 > 7 > 6 > 5 > 4 > 3 > 2 > 1 > +1.61E E E E E E E E E E-02 POTENTIALS (MICROVOLT) 9 > 8 > 7 > +1.95E E E+00.31E>00 6 > +1.09E+00 5 > +8.78E-01 4 > +6.63E-01 -, > +4.49E-01 2 > 1 > +2.34E-Q E-02 (C) Figure 4. Isopotential contours when the source density is 10 microcoulomb/cnf. (a) The stimulation site is laterally located. The ground reference is the most medial-ventral element, (b) Same as a) but the ground is two rows lateral, (c) The stimulation site is more medial than in a) and b), and the ground is same as in b).

6 188 Computatiol Biomedicine Table 1. Potential contour areas where essential nuclei are stimulated. (Source charge density and electrode position are the same as in Figure 4b.) VCN DCN LVN MVN STT FN TB SO Solid 1 Solid 2 Solid 3 Solid 4 Solid 5 Solid 6 Solid 7 (Area) #9 #9 #9 n DISCUSSION The DCN lies in thefloorof the lateral recess and the VCN is covered by the middle cerebellar peduncle. They are invisible to the surgeons and can not be directly accessed. Electrically-evoked auditory brainstem responses (EABRs) indicate the activation of the auditory structures when an electrode is in place. However, the electric field in the tissue being stimulated can not be directly measured. A better understanding of the electric fields produced in the brainstem tissue is needed for improving the auditory stimulation without major side effects. Finite element alysis has proved to be an useful tool in this respect. There are many factors affecting the electric field in the brainstem, such as changes of stimulation sites, tissue parameters, stimulation modes, ground reference positioning, biphasic sigl frequency and amplitude ranges, as well as the variations of the electrode shape and number. These factors work to shift the electric fields present in auditory and other surrounding structures. This model demonstrates the influence of reference electrode positions and sigl levels. The more medially the electrode is placed in the lateral recess, the larger the area of CNC stimulated, and the lower the level of sigl to other nouditory nuclei. Further 3D reconstruction and finite element modeling will be used to study different surface implantation sites, the effects of different modes of stimulation, and the electric fields produced by penetrating electrodes. REFERENCES Brackmann, D.E., Hitselberger, W.E., Nelson, R.A., Moore, J.K., Waring, M., Portillo, F., Shannon, R.V., and Telischi, F.F. * Auditory brainstem implant: I. issues in surgical implantation', ORL-HNS, in press, 1993.

7 Computatiol Biomedicine 189 Eisenberg, L.S., Maltan, A.A., Portillo, F., Mobley, J.P., and House, W.F. 'Electrical stimulation of the auditory brain stem structure in deafened adults', J. Reh. Res. Devlp., 24(3): 9-22, Finley, C.C., Wilson, B.S., and White, M.W. 'Models of neural responsiveness to electrical stimulation', In J.M. Miller and F.A. Spelman (Eds): Cochlear Implant: Models of the electrically stimulated ear, Springer-Verlag New York Inc., New York, 55, Geddes, L. A. and Baker, L.E. 'The specific resistance of biological material - a compendium of data for the biomedical engineer and physiologist', Med.&Biol. Engng., vol.5: Pergamon Press, Huang, J, Mobley, J.P., and Moore, J.K. 'Three dimensiol brainstem reconstruction and interpretation for human auditory implant' Proc. 2nd Intl. Conf. Med. Biol. Impl. Tech., Coventry, UK, McCreery, D.B., Agnew, W.F., Yuen, T.G., and Bullara, L. 'Charge density and charge per phase as cofactors in neural injury induced by electrical stimulation', IEEE Tran. Biom. Eng., 37(10): , Miller, C.E. and Henriquez, C.S. 'Finite element alysis of bioelectric phenome'. Crit. Re. Biom. Eng., 18(3): , Sepulveda, N.G., Walker, C.F., and Heath, R.G. 'Finite element alysis of current pathways with implanted electrodes', J. Biomed. Eng., Vol.5: 41-48, Shannon, R.V. 'A model of safe levels for electrical stimulation', IEEE Tran. Biom. Eng., 39(4): , Spilker, R.L. and Simon, B.R.(Ed). Computatiol methods in bioengineering. Am. Soc. Mech. Engr., New York, ACKNOWLEDGEMENT The authors wish to thank Dr. Jean Moore and Bryan Wu for the supply of histologic data, and Geraldine Leo for the help on the manuscript. This work is supported by NIDCD NO1-DC-2400, to the Huntington Memorial Research Institute, and the House Ear Institute.

Auditory and Vestibular Systems

Auditory and Vestibular Systems Objective To learn the functional organization of the auditory and vestibular systems To understand how one can use changes in auditory function following injury to localize