Hiroshima City University. Hiroshima, 731-3194 Japan. Hiroshima, 731-3194 Japan. Hiroshima, 731-3194 Japan. Abstract-Transcranial magnetic stimulation ...
Proceedings of the 2005 IEEE Engineering in Medicine and Biology 27th Annual Conference Shanghai, China, September 1-4, 2005
A Method for Accurate Estimation of Stimulating Sites Based on Cerebral Cortex Structure in Transcranial Magnetic Stimulation O. Hiwaki
T. Inoue
Hiroshima City University Hiroshima, 731-3194 Japan
Abstract-Transcranial magnetic stimulation (TMS) is a non-invasive and painless method to stimulate the cortex. Accurate estimation of stimulating points is much important to evaluate cortical functions such as cortical motor output maps. However, most studies using TMS have not achieved to estimate the stimulating points exactly. This study proposed the novel method to provide the accurate estimation of stimulating points of the cortex in TMS taking account of the individual cortical anatomy and property of nerve excitation. We conducted TMS mapping of the motor cortex to confirm the validity of this method. The cortex sites which innervate muscles of the upper limb and hand were successfully delineated in primary motor area.
I. INTRODUCTION Transcranial magnetic stimulation (TMS) [1] is an only method for noninvasive stimulation of cerebral cortex. TMS has contributed to clinical and basic research of brain function. Furthermore, its availability for therapy in psychiatric and neurological disorders is being investigated [2]. A figure-of-eight coil [3], which is commonly used as a stimulating coil in TMS, provides stimulation of a somewhat small area in the cortex with concentration of induced electric field at the point below the midpoint of the figure-of-eight coil. The point, at which the perpendicular line through the midpoint of the figure-of-eight coil intersects the surface of cortex, has been defined as the stimulating site in the mapping studies of human motor cortex in TMS [4]-[6]. However, the stimulating site determined in such a way is not supposed to be plausible because electric field induced in TMS is dispersed over the brain vectorially in contrast to the localized stimulation with an electrode in electrical stimulation. Simulation studies with nerve model revealed that the induced electric field parallel to the longitudinal axis of the axon is effective for the stimulation and stimulation occurs at the end of the axon [7][8]. Pyramidal neurons which are the input and output neurons in the cortex are aligned perpendicular to the cortical surface [9]. On the basis of these understandings, it seems reasonable to suppose that the site on a cortex with a great component of the induced electric field perpendicular to the cortical surface is the stimulating site where nerve excitation readily occurs in TMS.
0-7803-8740-6/05/$20.00 ©2005 IEEE.
Y. Wada
Hiroshima City University Hiroshima, 731-3194 Japan
Hiroshima City University Hiroshima, 731-3194 Japan
In this study, we proposed the novel method for accurate estimation of the stimulating site in TMS using the information of the anatomical brain structure obtained from MRI data. II. METHOD FOR ESTIMATION OF STIMULATING SITE IN TMS The basic unit of the cerebral cortex is the cylindrical column containing pyramidal neurons perpendicular to the cortical surface [9]. The model analyses revealed that a pyramidal neuron with a soma and a long straight axon generates the excitation at the point between the soma and the axon when the neuron is stimulated with electric field induced by pulsed magnetic field [7][8]. The simulation studies also showed that the induced electric field parallel to the straight axon is most effective for the nerve excitation. According to these understandings, the relative intensity of nerve stimulation in TMS corresponds to the strength of the component of the induced electrical field perpendicular to the cortical surface. Therefore, we estimate the stimulating site in the way as the following steps: A. Registering the positions of the subject’s head and the stimulating coil using a position sensor detecting optical markers at the moment of the magnetic stimulation. B. Calculating the distribution of the induced electric field on the surface of the brain and computing the component of the induced electric field perpendicular to the cortical surface at each grid on the cortical surface. A. Registering the positions of the head and the stimulating coil We used Polaris system (Northern Digital Inc., Ontario, Canada) which is an optical measurement system measuring the 3D positions of wireless markers. The position sensor of Polaris system detected retro-reflective optical markers and transferred the result via a serial interface to the host computer. The 3D position of the coil was calculated with the positions of the four sphere-shaped 11.5 mm-diameter markers which were fixed on the figure-of-eight coil. Five round-shaped sheet markers with 11.5 mm diameter were put on the subject’s face: three on the forehead, one on the nasion and one on the top of the nose (Fig. 1). To achieve this technique
5302
Fig.1. Subject set-up for TMS stimulation. Sphere-shaped markers mounted on the figure-of-eight coil for tracking the position of the coil. Sheet markers put on the face are used for measuring head position.
Fig.2. Method to calculate the vector nearly perpendicular to the cortex surface. The sum of the unit vector directed from the
of estimating the stimulating site with high accuracy, the MRI data with the actual subject is adapted. The skin surface of the MRI data was fitted on the actual positions of the round-shaped sheet markers on the subject’s face, and the brain coordinates in the MRI scans were transferred to the head and coil coordinates measured by Polaris system. In the resulting common coordinate system, the coil position is determined in relation to the cortex.
objective pixel to the surround pixels which do not represent cortex gives the vector nearly perpendicular to the cortex surface.
B. Calculating the distribution of component of the induced electric field perpendicular to the cortical surface The alignment of T1-weighted MR scans resulted in the 3D brain image which was composed with isotropic voxels of 2 × 2 × 2 mm size. The vector which is nearly perpendicular to the cortex surface at each pixel representing cortex surface was given by the sum of the unit vectors directed to the surrounding pixels which do not represent cortex (Fig.2). We calculated the distribution of induced electric field on the surface of cortex using the information of the coil position in relation to the cortex. The component of the electric field perpendicular to the cortical surface at each pixel representing the cortex surface was calculated with computation of the dot product of the electric field and the unit vector perpendicular to the cortex surface (Fig.3). The strength of the component of the electric field perpendicular to the cortical surface was regarded as the stimulating strength of each pixel of the cortical surface in TMS. III. EXPERIMENTAL RESULTS We stimulated the right side of the subject’s head with a figure-of-eight coil connected to the magnetic stimulator (Magstim 200, The Magstim Company Ltd, South West Wales, U.K.), and motor evoked potentials (MEPs) of the left upper limb were measured. The electrodes were put on Biceps, Triceps, Flexor Carpi Ulnaris (FCU), Brachioradialis(BR), Abductor Pollicis Brevis(APB) and Abductor Digiti Minimi (ADM) muscles.
Fig.3. The component of the induced electric field perpendicular to the cortical surface given by the dot product of the induced electric field and the unit vector perpendicular to the cortical surface.
The results of the estimation of the stimulating sites using our method were shown in Fig.4. The largest amplitudes of MEPs among 6 muscles were observed at Abductor Digiti Minimi muscle of the hand in case (A), Flexor Carpi Ulnaris muscle of the forearm in case (B) and Biceps muscle of the upper arm in case (C) shown in Fig.4 respectively. The strongest stimulating points were estimated in the primary motor cortex in all cases shown in Fig.4. The strongest stimulating point was located at the uppermost in case (C), and that was located at the lowermost in case (A). IV. DISCUSSION In this study, we proposed new method to estimate the stimulating points accurately in TMS. This method is based on the property of the nerve excitation in the magnetic
5303
Fig.4. (Left Side) The stimulating sites estimated in three cases of TMS. (Right Side) The amplitudes of MEPs corresponding to each case.
5304
stimulation and the anatomical structure of the cortex. In other words, we hypothesized that the pyramidal neurons aligned perpendicular to the cortical surface were stimulated easily with the induced electric field perpendicular to the cortical surface in TMS. We carried out the measurement of MEPs in TMS, and estimated the stimulating sites of the cortex using our proposed method. The stimulating site was estimated in the primary motor cortex whenever any MEPs were observed at the 6 muscles of the upper limb. It is known that the signals descending to the muscles come from the primary motor cortex, so the result showing the estimated stimulating sites located in the primary motor cortex indicates the validity of the proposed method. The strongest stimulating point was estimated at the uppermost point when the largest amplitude of MEP was observed at upper arm muscle, whereas that was estimated at the lowermost point when the largest amplitude of MEP was observed at the hand muscle. These observations coincide with the motor representation in order of the upper arm, forearm, hand from up to down. The stimulating sites were estimated in the so-called hand-knob area innervating the hand muscle when the MEPs were observed at the hand muscle such as the Abductor Pollicis Brevis muscle or Abductor Digiti Minimi muscle. In many cases of the experiments in TMS, MEPs were observed at plural muscles simultaneously. The neuronal clusters which synapse to the motor neurons innervating same muscle were tangled in the primary motor cortex [10]. Based on these understandings, it is supposed to be reasonable that plural MEPs were elicited by single stimulation of the primary motor cortex. The analysis of the relation between the amplitude of each MEP and the estimated stimulating sites may reveal the overlapped sites innervating respective muscles.
stimulation coregistered with cortical surfaces derived from magnetic resonance images,” J. Clin. Neurophysiol., vol. 15, pp344-350, 1998. [5] B. Boroojerdi, H. Foltys, T. Krings, U. Spetzger, A. Thron, and R.Topper, “Localization of the motor hand area using transcranial magnetic stimulation and functional magnetic resonance imaging,” Clin. Neurophysiol., vol. 110, pp. 699-704, 1999. [6] G. J. Ettinger, M. E. Leventon, W. E. Grimson, R. Kikinis, L. Gugino, W. Cote, L. Sprung, L . Aglio, M. E. Shenton, G. Potts, V. L. Hernandez, E. Alexander, “Experimentation with a transcranial magnetic stimulation system for functional brain mapping.” Med. Image Anal., vol.2, pp. 133-142, 1998. [7]
S. S. Nagarajan, M. Durand, and E. N. Warman, “Effects of Induced Electric Fields on Finite Neuronal Structures: A Simulation Study,” IEEE Trans. Biomed. Eng., vol. 40, pp.1175-1188, 1993.
[8] F. Rattay, “The Basic Mechanism for the Electrical Stimulation of the Nervous System,” Neuroscience, vol.89, pp.335-346, 1999 [9] V. B. Mountcastle, “The columnar organization of the neocortex,” Brain, vol. 120, pp.701-722, 1997. [10] C. G. Phillips and R. Porter, “Corticospinal neurons: their role in movement, Academic Press, London (1977)
V. CONCLUSIONS We proposed the new method to estimate the stimulating site precisely in TMS. The validity of this method was confirmed with the experiment of MEP measurement elicited by TMS.
REFERENCES [1] A. T. Barker, R. Jalinous, and I. L. Freeston, “Noninvasive magnetic stimulation of human motor cortex,” Lancet, vol.1, pp.1106-1107, 1985. [2] M. S. George, S. H. Lisanby, H. A. Sackeim, “Transcranial magnetic stimulation: applications in neuropsychiatry,” Archives of General Psychiatry, Vol.56, pp.300-311, 1999. [3] S. Ueno, T. Tashiro, and K. Harada, “Localized stimulation of neural tissues in the brain by means of paired configuration of time-varying magnetic fileds,” J. Appl. Phys., vol.64, pp.5862-5864, 1988. [4] G. F. Potts, L.D. Gugino, M. E. Leventon, W. E. Grimson, R. Kikinis, W. Cote, E. Alexander, J. E. Anderson, G.. J. Ettinger, L. S. Aglio, and M.E., “Shenton.Visual hemifield mapping using transcranial magnetic
5305
