J Med Syst (2014) 38:64 DOI 10.1007/s10916-014-0064-7
SYSTEMS-LEVEL QUALITY IMPROVEMENT
Design of a Dynamic Transcranial Magnetic Stimulation Coil System Sheng Ge & Ruoli Jiang & Ruimin Wang & Ji Chen
Received: 18 November 2013 / Accepted: 26 May 2014 / Published online: 24 June 2014 # Springer Science+Business Media New York 2014
Abstract To study the brain activity at the whole-head range, transcranial magnetic stimulation (TMS) researchers need to investigate brain activity over the whole head at multiple locations. In the past, this has been accomplished with multiple single TMS coils that achieve quasi whole-head array stimulation. However, these designs have low resolution and are difficult to position and control over the skull. In this study, we propose a new dynamic whole-head TMS mesh coil system. This system was constructed using several sagittal and coronal directional wires. Using both simulation and real experimental data, we show that by varying the current direction and strength of each wire, this new coil system can form both circular coils or figure-eight coils that have the same features as traditional TMS coils. Further, our new system is superior to current coil systems because stimulation parameters such as size, type, location, and timing of stimulation can be dynamically controlled within a single experiment.
Keyword Transcranial magnetic stimulation . Electromagnetic modeling . Magnetic field distribution
This article is part of the Topical Collection on Systems-Level Quality Improvement Sheng Ge and Ruoli Jiang contributed equally to this work. S. Ge (*) Key Laboratory of Child Development and Learning Science of Ministry of Education, Research Center for Learning Science, Southeast University, Nanjing, Jiangsu 210096, China e-mail: [email protected] R. Jiang : J. Chen Department of Electrical and Computer Engineering, University of Houston, Houston, TX 77204-4005, USA R. Wang : J. Chen School of Electronic Engineering and Optoelectronic Technology, Nanjing University of Science and Technology, Nanjing, Jiangsu 210094, China
Introduction Almost all human behaviors are based on complex interactions among different brain regions. To study how the different brain regions process information and interact with each other, it is necessary to investigate both spatial (where) and temporal (when) aspects of brain information processing. Techniques to passively capture brain signals, such as functional magnetic resonance imaging (fMRI) [1, 2], electroencephalography (EEG) [3, 4], magnetoencephalography (MEG) [5, 6], and near-infrared spectroscopy (NIRS) [7] can ascertain where and when activation occurs. However, they cannot determine whether a specific activation is necessary for a given task. With the accelerating development of brain science, passive imaging methods alone no longer satisfy the requirements of current neurological studies. In particular, a demand now exists for controlled experiments in which the brain is actively stimulated and observations can be made regarding how the brain interacts and processes information. Transcranial magnetic stimulation (TMS) is a non-invasive approach that meets such needs. TMS uses electromagnetic induction to generate an electric current across the scalp and skull without physical contact. A TMS coil produces a magnetic field oriented orthogonally to the plane of the coil. The magnetic field passes unimpeded through the skin and skull, inducing an oppositely directed current in the brain that activates nearby nerve cells in much the same way that a current applied directly to the cortical surface does. These currents can depolarize or hyperpolarize neurons in the brain [8], depending on the frequency of stimulation [9, 10]. As a non-invasive and effective method of making reversible lesions (“virtual lesion”) in the human brain [11], TMS is a noninvasive technique for transiently modulating neural activity. In cognitive neuroscience it is generally used with the objective of disrupting neural activity associated with cognitive processes, thus revealing the necessity of cortical regions in cognitive functions. Since the TMS stimulation position and timing can be controlled, TMS can be used to investigate not only the
64, Page 2 of 6
spatial but also the temporal characteristics of brain activation [12]. Current TMS coils can be classified as circular or figure-eight coils [9]. Circular coils induce broadly distributed eddy currents in the brain, while figure-eight coils consist of two circular windings in which currents flow in opposite directions and provide high eddycurrent density at the junction point of the two circular windings [13], enabling stimulation that is highly localized compared with that of circular coils [14]. However, single TMS coils currently available can only stimulate a single and fixed brain site. Studying interactions between brain regions is becoming an increasingly important aspect of neuroscience research. To study the brain interaction in a whole-head range, researchers need to investigate brain activity over the whole head at multiple locations, which is difficult to do with the current TMS coils. To compensate for this limitation, researchers have tried designing multiple TMS coils to achieve quasi whole-head stimulation [15–22] (Fig. 1). However, those designs still involve arranging multiple single coils over the skull. Because of the limitations of the shapes of the coils themselves and the interval between the coil elements, such designs preclude high-resolution configuration of coils. Moreover, to ensure the accurate positioning of coils over the skull, and the relative positions among individual coils, the operator needs to locate the coils accurately one by one. The difficulty in controlling the relative displacement of each coil makes it nearly impossible to repeat an experimental stimulation at the exact same location. Further, each single coil requires a separate control unit; a multi-channel system thus inevitably leads to a complicated control problem. Because of these issues, experimental Fig. 1 Past quasi whole-head array TMS coil systems. a Ref. [15–17]. b Ref.[18]. c Ref. [19]. d Ref. [20]. e Ref. [21]. f Ref. [22]
J Med Syst (2014) 38:64
designs using quasi whole-head array TMS coils have not been widely used in the study of dynamic brain activity. In this study, we propose a novel design employing a dynamic whole-head TMS mesh coil system to overcome the disadvantages of the current designs. In doing so, we have attempted to develop a true whole-head TMS mesh coil system. This system allows the researchers to stimulate multiple brain areas in a single experiment. In addition, it is easy to generate different stimulus patterns spatially and temporally. With this proposed TMS coil system, researchers will be able to reproducibly stimulate multiple sites with any operatorprogrammable spatial–temporal pattern, and thus investigate the spatiotemporal features of information processing in the brain. TMS studies with such a coil will increase our understanding of behavior as well as neurological and psychiatric pathologies.
Coil design and simulations Multi-channel mesh coil For the proposed system, a multi-channel mesh coil system as shown in Fig. 2 was developed. The coil system consists of a set of perpendicular wires aligned in the S and C directions, which correspond to the sagittal and coronal axes, respectively. Wires in the S and C directions physically cross each other but are electrically insulated. By varying the current direction in each wire, we can configure the proposed coil system into circular coils or figure-eight coils of different sizes (Fig. 3a and b). Letting current strengths S2=I, S3=−I, C7=I, and C8=−I, the
J Med Syst (2014) 38:64
Page 3 of 6, 64
Fig. 2 Outline of the proposed mesh coil system. a The proposed coil system comprises a set of wires in sagittal and coronal directions. b The proposed coil system is arranged along the surface of the skull. S: sagittal axis; C: coronal axis
proposed array coil can be configured as a small circular coil (Fig. 3a, top-left). Alternatively, by letting S4 =I, S8=−I, C2=I, and C6=−I, the proposed array coil can be configured as a large circular coil (Fig. 3a, bottomright). Similarly, this system allows configuration of figure-eight coils of different sizes (Fig. 3b). This design provides maximum flexibility, allowing the location and range of stimulation to the brain surface to be modifiable depending on experimental requirements. We used the Ansoft Maxwell software package (ANSYS, Inc., Canonsburg, PA, USA) to simulate the current density distribution for the traditional TMS coil and our proposed TMS coil. Figure 4 shows the simulation results for the magnetic field and current density distributions above a traditional figure-eight coil. A strong magnetic field was located inside each of the two circular windings (Fig. 4a), and the current peaked where the two coils joined, at a value that was about three times that found in other areas of the coils
Fig. 3 Proposed coil system generating different type coils of different sizes. a Circular coils of different sizes. b Figure-eight coils of different sizes
(Fig. 4b). These distributions are consistent with the theoretical characteristics of the traditional figure-eight coil [9, 13]. Figure 5 shows the simulation of the current density distribution in a plane 0.76 cm above the proposed figureeight mesh coil. The magnetic field generated by the wiremesh coils can be computed as follows. Consider a set of crossed wires. The magnetic field peaks at the crossing point of two wires. From the integral form of Ampere’s Law, ! ! ∮c B ⋅d l ¼ μ0 I;
ð1Þ
We can calculate the magnetic field B generated by a single wire around closed path C. Here, I is the current flow in the wire and μ0 is the magnetic constant. In the reference medium of a classical vacuum, μ0 =4π×10−7H⋅m−1. If C is a circular path with radius h centered at the wire and oriented perpendicular to it, the amplitude B of
64, Page 4 of 6
J Med Syst (2014) 38:64
Figure 5 shows that while there is no peak in the magnetic field where C2 and S2 cross, the magnetic field is relatively strong at other crossings. This type of current-density distribution is undesirable because the ideal figure-eight coil should generate peak current-density at the junction of two neighboring loop coils (where C2 crosses S2). This undesirable distribution occurs because in addition to flowing through the figure-eight loop, current also flows through the other sections of the feed wires. These ‘byproduct’ currents may generate current densities that affect the performance of the wire-mesh system. Two-layer mesh coil
Fig. 4 Simulation results of a traditional figure-eight coil. a The magnetic field distribution. b The current density distribution
the magnetic field will be uniform along path C and is given by: B¼
μ0 I ; 2πh
ð2Þ
where h is the distance from the wire. For two crossed wires, the net magnetic field Bs is the vector sum of the fields generated by each of them. If two wires cross at a right angle and the diameters of the wires are ignored, the amplitude of the magnetic field at distance h on top of the crossing point is: pffiffiffi 2μ0 I Bs ¼ : ð3Þ 2πh
Fig. 5 Simulation of the current density distribution of a figure-eight mesh coil
To overcome the issue in the design of figure-eight coil described above, we designed two layers of mesh coil to cancel out the effects caused by current flowing through the feed wires. The basic idea of the two-layer mesh coil is to set one mesh coil above a second mesh coil, in which the first layer (Layer 1; Fig. 6a, left) has the same design as depicted in Fig. 3b, and the second layer (Layer 2; Fig. 6a, middle) has a different design that eliminates the current densities induced by the peripheral currents in first layer. Using this two-layer mesh coil, the currents in the second layer that flow outside of the figure-eight coil oppose those in the first layer, resulting in an ideal net current-density of zero in the feed wires and a doubled current density in the figure-eight coil (Fig. 6a, right). In the case of the circular coil design, the first-layer (Layer 1; Fig. 6b, left) mesh coil design is the same as that shown in Fig. 3a, while the configuration of the second layer (Layer 2; Fig. 6b, middle) is designed with currents flowing in the same direction. This results in a net current-density for the two-layer
Fig. 6 Two-layer mesh coil configuration. a Two-layer figure-eight coil. b Two-layer circular coil
J Med Syst (2014) 38:64
circular mesh coil (Fig. 6b, right) that is twice that of the single-layer design depicted in Fig. 3a. Figure 7 shows the simulation results of the current density distribution in a plane 0.76 cm above the redesigned two-layer figure-eight coil. Unlike the case shown in Fig. 5, a peak value was observed only where C2 and S2 cross, and not at other wire crossings. This type of current-density distribution is qualitatively similar to that generated by a traditional figure-eight coil (Fig. 4b). Thus, it is considered that such a two-layer mesh coil can realize the current density distribution that is qualitatively similar to that of the traditional TMS coil.
Page 5 of 6, 64
Fig. 8 Magnetic field (dB/dt) ) measured 1 cm above the figure-eight coil that was configured using the two-layer mesh coil
Dynamic TMS stimulus generation Bench experiment To determine whether our simulation results translate into real stimulation, we also conducted experimental tests. Because of hardware limitations, we cannot presently make a direct measurement of induced currents within a phantom. However, because the magnetic field generated by the coil is directly related to the induced current within the targeted brain region, the observed magnetic field pattern also reveals the underlying distribution of induced currents. Therefore, we measured the magnetic field (dB/dt) generated by a figure-eight coil that was configured using the two-layer mesh coil. Based on the measurement results, it was clear that strong magnetic fields were generated within each of the two circular windings, which is qualitatively similar to the simulated distribution of a traditional figure-eight coil (Fig. 8, see Fig. 4a for comparison). From this result, we conclude that our proposed two-layer mesh coil can generate a magnetic field that is qualitatively similar to that of the traditional TMS coil.
Fig. 7 Simulation results showing the current-density distribution of a figure-eight coil using the reconfigured two-layer mesh coil
Figure 9 illustrates the technique of dynamic TMS in which current in different wires is simply switched on or off to generate different sizes and types of coil configurations at different time instances. For example, at time point T1 the configuration could be a figure-eight at the top-left corner. Then, at time T2 the configuration can easily be changed to a large circular coil at the bottom-left corner. Similarly, at time T3, the configuration can be changed again to create two small circular coils at the bottom. Thus, by varying the current direction in each wire, the type and location of the TMS coil can be changed at different time points. By pre-storing sequences of stimulation parameters within a control unit, our proposed mesh coil can be used to create dynamic TMSstimuli that differ in location, size, and type.
Conclusion In this study, we propose a dynamic whole-head TMS mesh coil system. Our new design has several distinct advantages. First, the mesh coil system reduces relative movement compared with quasi-array coil designs that have been proposed [15–22]. In quasi-array coil designs, the position of each single coil element needs to be set separately, and the relative
Fig. 9 Illustration of a potential dynamic procedure using the proposed reconfigurable coil. At time T1, a figure-eight is configured at the top-left corner; At time T2, a large circular coil is configured at the bottom-left corner; At time T3, two small circular coils are configured at the bottom
64, Page 6 of 6
displacement among the coils may cause offsets of the magnetic field. Our proposed mesh coil system has a fixed shape, which can eliminate the relative displacement among individual coils. Second, it is highly flexible in terms of the sizes and types of stimulation. By varying the current direction and strength in each wire, we can configure the mesh coil into circular and figure-eight coils of different sizes. Third, the proposed coil system has superior flexibility in terms of location, range, and timing of stimulation, which can be modified to satisfy specific experimental requirements. Using preprogrammed stimulation patterns and ranges within the control unit, the system can be used with different stimulation configurations within a single experimental trial. Fourth, the proposed coil system requires vastly fewer control units. Quasi N-by-N coil arrays need N×N control units, while our proposed design needs only 2×(N+1) control units for an Nby-N coil array. This design will therefore simplify the complex control system of large coil-array systems. Acknowledgments This work was supported by the National Natural Science Foundation of China (No. 51007040).
References 1. Ogawa, S., Tank, D. W., Menon, R., Ellermann, J. M., Kim, S. G., et al., Intrinsic signal changes accompanying sensory stimulation: Functional brain mapping with magnetic resonance imaging. Proc. Natl. Acad. Sci. U. S. A. 89:5951–5955, 1992. 2. Lin, P., Hasson, U., Jovicich, J., and Robinson, S., A neuronal basis for task-negative responses in the human brain. Cereb. Cortex 21: 821–830, 2011. 3. Teplan, M., Fundamentals of EEG measurement. Meas. Sci. Rev. 2: 1–11, 2002. 4. Subha, D. P., Joseph, P. K., Acharya, R., and Lim, C. M., EEG signal analysis: A survey. J. Med. Syst. 34:195–212, 2010. 5. Cohen, D., Magnetoencephalography: Detection of the brain’s electrical activity with a superconducting magnetometer. Science 175: 664–666, 1972. 6. Hari, R., and Salmelin, R., Magnetoencephalography: From SQUIDs to neuroscience: Neuroimage 20th anniversary special edition. Neuroimage 61:386–396, 2012. 7. Villringer, A., Planck, J., Hock, C., Schleinkofer, L., and Dirnagl, U., Near infrared spectroscopy (NIRS): A new tool to study hemodynamic changes during activation of brain function in human adults. Neurosci. Lett. 154:101–104, 1993.
J Med Syst (2014) 38:64 8. Oliveri, M., Caltagirone, C., Filippi, M. M., Traversa, R., Cicinelli, P., et al., Paired transcranial magnetic stimulation protocols reveal a pattern of inhibition and facilitation in the human parietal cortex. J. Physiol. 529:461–468, 2000. 9. Ueno, S., Tashiro, T., and Harada, K., Localized stimulation of neural tissues in the brain by means of a paired configuration of time-varying magnetic fields. J. Appl. Phys. 64: 5862–5864, 1988. 10. Cacioppo, J. T., Tassinary, L. G., and Berntson, G. G., Handbook of psychophysiology, 3rd edition. Cambridge University Press, New York, 2007. 11. Walsh, V., and Cowey, A., Transcranial magnetic stimulation and cognitive neuroscience. Nat. Rev. Neurosci. 1:73–79, 2000. 12. Kobayashi, M., and Pascual-Leone, A., Transcranial magnetic stimulation in neurology. Lancet Neurol. 2:145–156, 2003. 13. Bohning, D. E., Introduction and overview of TMS physics. In: George, M. S., and Belmaker, R. H. (Eds.), Transcranial magnetic stimulation in neuropsychiatry. American Psychiatric Press, Washington, DC, pp. 13–44, 2000. 14. Burunkaya, M., Design and construction of a low cost dsPIC controller based repetitive transcranial magnetic stimulator (rTMS). J. Med. Syst. 34:15–24, 2010. 15. Ruohonen, J., and Ilmoniemi, R. J., Multichannel magnetic stimulation: Improved stimulus targeting. In: Nilsson, J., Panizza, M., and Grandori, F. (Eds.), Advances in magnetic stimulation: mathematical modeling and clinical applications. Maugeri Foundation, Pavia, pp. 55–64, 1996. 16. Ruohonen, J., and Ilmoniemi, R. J., Focusing and targeting of magnetic brain stimulation using multiple coils. Med. Biol. Eng. Comput. 36:297–301, 1998. 17. Ruohonen, J., Ravazzani, P., Grandori, F., and Ilmoniemi, R. J., Theory of multichannel magnetic stimulation: Toward functional neuromuscular rehabilitation. IEEE Trans. Biomed. Eng. 46:646– 651, 1999. 18. Ho, S. L., Xu, G. Z., Fu, W. N., Yang, Q. X., Hou, H. J., et al., Optimization of array magnetic coil design for functional magnetic stimulation based on improved genetic algorithm. IEEE Trans. Magn. 45:4849–4852, 2009. 19. Han, B. H., Chun, I. K., Lee, S. C., and Lee, S. Y., Multichannel magnetic stimulation system design considering mutual couplings among the stimulation coils. IEEE Trans. Biomed. Eng. 51:812–817, 2004. 20. Wang, X. M., Chen, Y., Guo, M. X., and Wang, M. S., Design of multi-channel brain magnetic stimulator and ANSYS simulation. IJBEM 7:259–262, 2005. 21. Im, C. H., and Lee, C., Computer-aided performance evaluation of a multichannel transcranial magnetic stimulation system. IEEE Trans. Magn. 42:3803–3808, 2006. 22. Xu, G. Z., Chen, Y., Yang, S., Wang, M. S., Yan, W. L., The optimal design of magnetic coil in transcranial magnetic stimulation. 27th Annual International Conference of the IEEE Engineering in Medicine and Biology Society: 6221–6224, 2005
SYSTEMS-LEVEL QUALITY IMPROVEMENT
Design of a Dynamic Transcranial Magnetic Stimulation Coil System Sheng Ge & Ruoli Jiang & Ruimin Wang & Ji Chen
Received: 18 November 2013 / Accepted: 26 May 2014 / Published online: 24 June 2014 # Springer Science+Business Media New York 2014
Abstract To study the brain activity at the whole-head range, transcranial magnetic stimulation (TMS) researchers need to investigate brain activity over the whole head at multiple locations. In the past, this has been accomplished with multiple single TMS coils that achieve quasi whole-head array stimulation. However, these designs have low resolution and are difficult to position and control over the skull. In this study, we propose a new dynamic whole-head TMS mesh coil system. This system was constructed using several sagittal and coronal directional wires. Using both simulation and real experimental data, we show that by varying the current direction and strength of each wire, this new coil system can form both circular coils or figure-eight coils that have the same features as traditional TMS coils. Further, our new system is superior to current coil systems because stimulation parameters such as size, type, location, and timing of stimulation can be dynamically controlled within a single experiment.
Keyword Transcranial magnetic stimulation . Electromagnetic modeling . Magnetic field distribution
This article is part of the Topical Collection on Systems-Level Quality Improvement Sheng Ge and Ruoli Jiang contributed equally to this work. S. Ge (*) Key Laboratory of Child Development and Learning Science of Ministry of Education, Research Center for Learning Science, Southeast University, Nanjing, Jiangsu 210096, China e-mail: [email protected] R. Jiang : J. Chen Department of Electrical and Computer Engineering, University of Houston, Houston, TX 77204-4005, USA R. Wang : J. Chen School of Electronic Engineering and Optoelectronic Technology, Nanjing University of Science and Technology, Nanjing, Jiangsu 210094, China
Introduction Almost all human behaviors are based on complex interactions among different brain regions. To study how the different brain regions process information and interact with each other, it is necessary to investigate both spatial (where) and temporal (when) aspects of brain information processing. Techniques to passively capture brain signals, such as functional magnetic resonance imaging (fMRI) [1, 2], electroencephalography (EEG) [3, 4], magnetoencephalography (MEG) [5, 6], and near-infrared spectroscopy (NIRS) [7] can ascertain where and when activation occurs. However, they cannot determine whether a specific activation is necessary for a given task. With the accelerating development of brain science, passive imaging methods alone no longer satisfy the requirements of current neurological studies. In particular, a demand now exists for controlled experiments in which the brain is actively stimulated and observations can be made regarding how the brain interacts and processes information. Transcranial magnetic stimulation (TMS) is a non-invasive approach that meets such needs. TMS uses electromagnetic induction to generate an electric current across the scalp and skull without physical contact. A TMS coil produces a magnetic field oriented orthogonally to the plane of the coil. The magnetic field passes unimpeded through the skin and skull, inducing an oppositely directed current in the brain that activates nearby nerve cells in much the same way that a current applied directly to the cortical surface does. These currents can depolarize or hyperpolarize neurons in the brain [8], depending on the frequency of stimulation [9, 10]. As a non-invasive and effective method of making reversible lesions (“virtual lesion”) in the human brain [11], TMS is a noninvasive technique for transiently modulating neural activity. In cognitive neuroscience it is generally used with the objective of disrupting neural activity associated with cognitive processes, thus revealing the necessity of cortical regions in cognitive functions. Since the TMS stimulation position and timing can be controlled, TMS can be used to investigate not only the
64, Page 2 of 6
spatial but also the temporal characteristics of brain activation [12]. Current TMS coils can be classified as circular or figure-eight coils [9]. Circular coils induce broadly distributed eddy currents in the brain, while figure-eight coils consist of two circular windings in which currents flow in opposite directions and provide high eddycurrent density at the junction point of the two circular windings [13], enabling stimulation that is highly localized compared with that of circular coils [14]. However, single TMS coils currently available can only stimulate a single and fixed brain site. Studying interactions between brain regions is becoming an increasingly important aspect of neuroscience research. To study the brain interaction in a whole-head range, researchers need to investigate brain activity over the whole head at multiple locations, which is difficult to do with the current TMS coils. To compensate for this limitation, researchers have tried designing multiple TMS coils to achieve quasi whole-head stimulation [15–22] (Fig. 1). However, those designs still involve arranging multiple single coils over the skull. Because of the limitations of the shapes of the coils themselves and the interval between the coil elements, such designs preclude high-resolution configuration of coils. Moreover, to ensure the accurate positioning of coils over the skull, and the relative positions among individual coils, the operator needs to locate the coils accurately one by one. The difficulty in controlling the relative displacement of each coil makes it nearly impossible to repeat an experimental stimulation at the exact same location. Further, each single coil requires a separate control unit; a multi-channel system thus inevitably leads to a complicated control problem. Because of these issues, experimental Fig. 1 Past quasi whole-head array TMS coil systems. a Ref. [15–17]. b Ref.[18]. c Ref. [19]. d Ref. [20]. e Ref. [21]. f Ref. [22]
J Med Syst (2014) 38:64
designs using quasi whole-head array TMS coils have not been widely used in the study of dynamic brain activity. In this study, we propose a novel design employing a dynamic whole-head TMS mesh coil system to overcome the disadvantages of the current designs. In doing so, we have attempted to develop a true whole-head TMS mesh coil system. This system allows the researchers to stimulate multiple brain areas in a single experiment. In addition, it is easy to generate different stimulus patterns spatially and temporally. With this proposed TMS coil system, researchers will be able to reproducibly stimulate multiple sites with any operatorprogrammable spatial–temporal pattern, and thus investigate the spatiotemporal features of information processing in the brain. TMS studies with such a coil will increase our understanding of behavior as well as neurological and psychiatric pathologies.
Coil design and simulations Multi-channel mesh coil For the proposed system, a multi-channel mesh coil system as shown in Fig. 2 was developed. The coil system consists of a set of perpendicular wires aligned in the S and C directions, which correspond to the sagittal and coronal axes, respectively. Wires in the S and C directions physically cross each other but are electrically insulated. By varying the current direction in each wire, we can configure the proposed coil system into circular coils or figure-eight coils of different sizes (Fig. 3a and b). Letting current strengths S2=I, S3=−I, C7=I, and C8=−I, the
J Med Syst (2014) 38:64
Page 3 of 6, 64
Fig. 2 Outline of the proposed mesh coil system. a The proposed coil system comprises a set of wires in sagittal and coronal directions. b The proposed coil system is arranged along the surface of the skull. S: sagittal axis; C: coronal axis
proposed array coil can be configured as a small circular coil (Fig. 3a, top-left). Alternatively, by letting S4 =I, S8=−I, C2=I, and C6=−I, the proposed array coil can be configured as a large circular coil (Fig. 3a, bottomright). Similarly, this system allows configuration of figure-eight coils of different sizes (Fig. 3b). This design provides maximum flexibility, allowing the location and range of stimulation to the brain surface to be modifiable depending on experimental requirements. We used the Ansoft Maxwell software package (ANSYS, Inc., Canonsburg, PA, USA) to simulate the current density distribution for the traditional TMS coil and our proposed TMS coil. Figure 4 shows the simulation results for the magnetic field and current density distributions above a traditional figure-eight coil. A strong magnetic field was located inside each of the two circular windings (Fig. 4a), and the current peaked where the two coils joined, at a value that was about three times that found in other areas of the coils
Fig. 3 Proposed coil system generating different type coils of different sizes. a Circular coils of different sizes. b Figure-eight coils of different sizes
(Fig. 4b). These distributions are consistent with the theoretical characteristics of the traditional figure-eight coil [9, 13]. Figure 5 shows the simulation of the current density distribution in a plane 0.76 cm above the proposed figureeight mesh coil. The magnetic field generated by the wiremesh coils can be computed as follows. Consider a set of crossed wires. The magnetic field peaks at the crossing point of two wires. From the integral form of Ampere’s Law, ! ! ∮c B ⋅d l ¼ μ0 I;
ð1Þ
We can calculate the magnetic field B generated by a single wire around closed path C. Here, I is the current flow in the wire and μ0 is the magnetic constant. In the reference medium of a classical vacuum, μ0 =4π×10−7H⋅m−1. If C is a circular path with radius h centered at the wire and oriented perpendicular to it, the amplitude B of
64, Page 4 of 6
J Med Syst (2014) 38:64
Figure 5 shows that while there is no peak in the magnetic field where C2 and S2 cross, the magnetic field is relatively strong at other crossings. This type of current-density distribution is undesirable because the ideal figure-eight coil should generate peak current-density at the junction of two neighboring loop coils (where C2 crosses S2). This undesirable distribution occurs because in addition to flowing through the figure-eight loop, current also flows through the other sections of the feed wires. These ‘byproduct’ currents may generate current densities that affect the performance of the wire-mesh system. Two-layer mesh coil
Fig. 4 Simulation results of a traditional figure-eight coil. a The magnetic field distribution. b The current density distribution
the magnetic field will be uniform along path C and is given by: B¼
μ0 I ; 2πh
ð2Þ
where h is the distance from the wire. For two crossed wires, the net magnetic field Bs is the vector sum of the fields generated by each of them. If two wires cross at a right angle and the diameters of the wires are ignored, the amplitude of the magnetic field at distance h on top of the crossing point is: pffiffiffi 2μ0 I Bs ¼ : ð3Þ 2πh
Fig. 5 Simulation of the current density distribution of a figure-eight mesh coil
To overcome the issue in the design of figure-eight coil described above, we designed two layers of mesh coil to cancel out the effects caused by current flowing through the feed wires. The basic idea of the two-layer mesh coil is to set one mesh coil above a second mesh coil, in which the first layer (Layer 1; Fig. 6a, left) has the same design as depicted in Fig. 3b, and the second layer (Layer 2; Fig. 6a, middle) has a different design that eliminates the current densities induced by the peripheral currents in first layer. Using this two-layer mesh coil, the currents in the second layer that flow outside of the figure-eight coil oppose those in the first layer, resulting in an ideal net current-density of zero in the feed wires and a doubled current density in the figure-eight coil (Fig. 6a, right). In the case of the circular coil design, the first-layer (Layer 1; Fig. 6b, left) mesh coil design is the same as that shown in Fig. 3a, while the configuration of the second layer (Layer 2; Fig. 6b, middle) is designed with currents flowing in the same direction. This results in a net current-density for the two-layer
Fig. 6 Two-layer mesh coil configuration. a Two-layer figure-eight coil. b Two-layer circular coil
J Med Syst (2014) 38:64
circular mesh coil (Fig. 6b, right) that is twice that of the single-layer design depicted in Fig. 3a. Figure 7 shows the simulation results of the current density distribution in a plane 0.76 cm above the redesigned two-layer figure-eight coil. Unlike the case shown in Fig. 5, a peak value was observed only where C2 and S2 cross, and not at other wire crossings. This type of current-density distribution is qualitatively similar to that generated by a traditional figure-eight coil (Fig. 4b). Thus, it is considered that such a two-layer mesh coil can realize the current density distribution that is qualitatively similar to that of the traditional TMS coil.
Page 5 of 6, 64
Fig. 8 Magnetic field (dB/dt) ) measured 1 cm above the figure-eight coil that was configured using the two-layer mesh coil
Dynamic TMS stimulus generation Bench experiment To determine whether our simulation results translate into real stimulation, we also conducted experimental tests. Because of hardware limitations, we cannot presently make a direct measurement of induced currents within a phantom. However, because the magnetic field generated by the coil is directly related to the induced current within the targeted brain region, the observed magnetic field pattern also reveals the underlying distribution of induced currents. Therefore, we measured the magnetic field (dB/dt) generated by a figure-eight coil that was configured using the two-layer mesh coil. Based on the measurement results, it was clear that strong magnetic fields were generated within each of the two circular windings, which is qualitatively similar to the simulated distribution of a traditional figure-eight coil (Fig. 8, see Fig. 4a for comparison). From this result, we conclude that our proposed two-layer mesh coil can generate a magnetic field that is qualitatively similar to that of the traditional TMS coil.
Fig. 7 Simulation results showing the current-density distribution of a figure-eight coil using the reconfigured two-layer mesh coil
Figure 9 illustrates the technique of dynamic TMS in which current in different wires is simply switched on or off to generate different sizes and types of coil configurations at different time instances. For example, at time point T1 the configuration could be a figure-eight at the top-left corner. Then, at time T2 the configuration can easily be changed to a large circular coil at the bottom-left corner. Similarly, at time T3, the configuration can be changed again to create two small circular coils at the bottom. Thus, by varying the current direction in each wire, the type and location of the TMS coil can be changed at different time points. By pre-storing sequences of stimulation parameters within a control unit, our proposed mesh coil can be used to create dynamic TMSstimuli that differ in location, size, and type.
Conclusion In this study, we propose a dynamic whole-head TMS mesh coil system. Our new design has several distinct advantages. First, the mesh coil system reduces relative movement compared with quasi-array coil designs that have been proposed [15–22]. In quasi-array coil designs, the position of each single coil element needs to be set separately, and the relative
Fig. 9 Illustration of a potential dynamic procedure using the proposed reconfigurable coil. At time T1, a figure-eight is configured at the top-left corner; At time T2, a large circular coil is configured at the bottom-left corner; At time T3, two small circular coils are configured at the bottom
64, Page 6 of 6
displacement among the coils may cause offsets of the magnetic field. Our proposed mesh coil system has a fixed shape, which can eliminate the relative displacement among individual coils. Second, it is highly flexible in terms of the sizes and types of stimulation. By varying the current direction and strength in each wire, we can configure the mesh coil into circular and figure-eight coils of different sizes. Third, the proposed coil system has superior flexibility in terms of location, range, and timing of stimulation, which can be modified to satisfy specific experimental requirements. Using preprogrammed stimulation patterns and ranges within the control unit, the system can be used with different stimulation configurations within a single experimental trial. Fourth, the proposed coil system requires vastly fewer control units. Quasi N-by-N coil arrays need N×N control units, while our proposed design needs only 2×(N+1) control units for an Nby-N coil array. This design will therefore simplify the complex control system of large coil-array systems. Acknowledgments This work was supported by the National Natural Science Foundation of China (No. 51007040).
References 1. Ogawa, S., Tank, D. W., Menon, R., Ellermann, J. M., Kim, S. G., et al., Intrinsic signal changes accompanying sensory stimulation: Functional brain mapping with magnetic resonance imaging. Proc. Natl. Acad. Sci. U. S. A. 89:5951–5955, 1992. 2. Lin, P., Hasson, U., Jovicich, J., and Robinson, S., A neuronal basis for task-negative responses in the human brain. Cereb. Cortex 21: 821–830, 2011. 3. Teplan, M., Fundamentals of EEG measurement. Meas. Sci. Rev. 2: 1–11, 2002. 4. Subha, D. P., Joseph, P. K., Acharya, R., and Lim, C. M., EEG signal analysis: A survey. J. Med. Syst. 34:195–212, 2010. 5. Cohen, D., Magnetoencephalography: Detection of the brain’s electrical activity with a superconducting magnetometer. Science 175: 664–666, 1972. 6. Hari, R., and Salmelin, R., Magnetoencephalography: From SQUIDs to neuroscience: Neuroimage 20th anniversary special edition. Neuroimage 61:386–396, 2012. 7. Villringer, A., Planck, J., Hock, C., Schleinkofer, L., and Dirnagl, U., Near infrared spectroscopy (NIRS): A new tool to study hemodynamic changes during activation of brain function in human adults. Neurosci. Lett. 154:101–104, 1993.
J Med Syst (2014) 38:64 8. Oliveri, M., Caltagirone, C., Filippi, M. M., Traversa, R., Cicinelli, P., et al., Paired transcranial magnetic stimulation protocols reveal a pattern of inhibition and facilitation in the human parietal cortex. J. Physiol. 529:461–468, 2000. 9. Ueno, S., Tashiro, T., and Harada, K., Localized stimulation of neural tissues in the brain by means of a paired configuration of time-varying magnetic fields. J. Appl. Phys. 64: 5862–5864, 1988. 10. Cacioppo, J. T., Tassinary, L. G., and Berntson, G. G., Handbook of psychophysiology, 3rd edition. Cambridge University Press, New York, 2007. 11. Walsh, V., and Cowey, A., Transcranial magnetic stimulation and cognitive neuroscience. Nat. Rev. Neurosci. 1:73–79, 2000. 12. Kobayashi, M., and Pascual-Leone, A., Transcranial magnetic stimulation in neurology. Lancet Neurol. 2:145–156, 2003. 13. Bohning, D. E., Introduction and overview of TMS physics. In: George, M. S., and Belmaker, R. H. (Eds.), Transcranial magnetic stimulation in neuropsychiatry. American Psychiatric Press, Washington, DC, pp. 13–44, 2000. 14. Burunkaya, M., Design and construction of a low cost dsPIC controller based repetitive transcranial magnetic stimulator (rTMS). J. Med. Syst. 34:15–24, 2010. 15. Ruohonen, J., and Ilmoniemi, R. J., Multichannel magnetic stimulation: Improved stimulus targeting. In: Nilsson, J., Panizza, M., and Grandori, F. (Eds.), Advances in magnetic stimulation: mathematical modeling and clinical applications. Maugeri Foundation, Pavia, pp. 55–64, 1996. 16. Ruohonen, J., and Ilmoniemi, R. J., Focusing and targeting of magnetic brain stimulation using multiple coils. Med. Biol. Eng. Comput. 36:297–301, 1998. 17. Ruohonen, J., Ravazzani, P., Grandori, F., and Ilmoniemi, R. J., Theory of multichannel magnetic stimulation: Toward functional neuromuscular rehabilitation. IEEE Trans. Biomed. Eng. 46:646– 651, 1999. 18. Ho, S. L., Xu, G. Z., Fu, W. N., Yang, Q. X., Hou, H. J., et al., Optimization of array magnetic coil design for functional magnetic stimulation based on improved genetic algorithm. IEEE Trans. Magn. 45:4849–4852, 2009. 19. Han, B. H., Chun, I. K., Lee, S. C., and Lee, S. Y., Multichannel magnetic stimulation system design considering mutual couplings among the stimulation coils. IEEE Trans. Biomed. Eng. 51:812–817, 2004. 20. Wang, X. M., Chen, Y., Guo, M. X., and Wang, M. S., Design of multi-channel brain magnetic stimulator and ANSYS simulation. IJBEM 7:259–262, 2005. 21. Im, C. H., and Lee, C., Computer-aided performance evaluation of a multichannel transcranial magnetic stimulation system. IEEE Trans. Magn. 42:3803–3808, 2006. 22. Xu, G. Z., Chen, Y., Yang, S., Wang, M. S., Yan, W. L., The optimal design of magnetic coil in transcranial magnetic stimulation. 27th Annual International Conference of the IEEE Engineering in Medicine and Biology Society: 6221–6224, 2005
