Summary
Source localization in the brain remains an ill-posed problem unless further constraints about the type of sources and the head model are imposed. Human head is modeled in various ways depending critically on the computing power available and/or the required level of accuracy. Sophisticated and truly representative models may yield more accurate results in general, but at the cost of prohibitively long computer times and huge memory requirements. In conventional source localization techniques, solution source parameters are taken as those which minimize an index of performance, defined relative to the model-generated and clinically measured voltages. We propose the use of a neural network in the place of commonly employed minimization algorithms such as the Simplex Method and the Marquardt algorithm, which are iterative and time consuming. With the aid of the error-backpropagation technique, a neural network is trained to compute source parameters, starting from a voltage set measured on the scalp. Here we describe the methods of training the neural network and investigate its localization accuracy. Based on the results of extensive studies, we conclude that neural networks are highly feasible as source localizers. A trained neural network's independence of localization speed from the head model, and the rapid localization ability, makes it possible to employ the most complex head model with the ease of the simplest model. No initial parameters need to be guessed in order to start the calculation, implying a possible automation of the entire localization process. One may train the network on experimental data, if available, thereby possibly doing away with head models.
Similar content being viewed by others
References
Abeyratne U.R., Kinouchi Y., Oki H., Okada J., Shichijo F., and Matsumoto K. A New Dipole Localization Method Using a Backpropagation Neural Network, Brain Topography, 1990, 3(1): 221
Ary J.P., Klein S.A., and Fender D.H. Location of Sources of Evoked Scalp Potentials: Corrections for skull and Scalp Thicknesses- Vol BME-28, No. 06, 1981, pp 447–452.
Caceci M.S., and Cacheris W.P. Fitting Curves to Data, BYTE, 1984,9: pp 340–361.
Elman J. and Zisper D. Learning the Hidden Structure of Speech, UCSD Inst. of Cognitive Sciences, Rep 8701, Feb. 1987.
Gaumond R.P., Lin J.-H., and Geselowitz D.B. Accuracy of Dipole Localization with a Spherical Homogeneous Model, IEEE Tr. on Biomedical Engineering, Vol BME-30, 1983, pp 29–33.
Gevins A. Analysis of the Electromagnetic Signals of the Human Brain: Milestones, Obstacles and Goals, IEEE Tr. on Biomedical Engineering, Vol BME-31, No. 12, 1984, pp 833–551.
Kavanagh R.M., Darcey T.M, Lehmann D., and Fender D.H. Evaluation of Methods for Three Dimensional Localization of Electrical Sources in the Human Brain, IEEE Tr. on Biomedical Engineering, Vol BME-25, No. 05, 1978, pp 421–429.
Marquardt D. An Algorithm for Least Squares Estimation of non-linear Parameters, SIAM J.Appl.Math. 11, 1963, pp 431–441.
Nunez P.L. Electrical Fields of the Brain: The Neurophysics of EEG, Oxford University Press, New York, 1981.
Nunez P.L. The Brain's Magnetic Field: Some effects of Multiple Sources on Localization methods, Electroenceph. and Clin. Neurophysiol., 1986, 63: 75–82.
Nunez P.L. Estimation of Large Neurocortical Source Activity with EEG Surface Laplacians, Brain Topography, 1989, 2 (1/2): 141–153.
Nuwer M.R. Quantitative EEG: I Techniques and Problems of Frequency analysis and Topographic Mapping, Journal of Clinical Neurophysiology, 5(1): 1–43, 1988.
Pool K.D., Aronofsky J.S., Finitzo T., and Barr R.S. Localization of Multiple Dipoles, Mathematical Programming Approaches, Brain Topography, 1989, 1 (4): 247–255.
Regan D. Human Brain Electrophysiology, Elsvier Sci. Publishing Co., 1989.
Rumelhart D.E., and McClellend J.L. Parallel Distributed Processing, MIT Press, 1986.
Rush S., and Driscoll D.A. Current distribution in the Brain from Surface Electrodes, Anesthesia and Analgesia 47, pp 717–723, 1968.
Scherg M. Fundamentals of Dipole Source Potential Analysis, In Auditory Evoked Magnetic Fields and Electric Potentials", Adv, Audiol. Basel, Karger, 1990, Vol 6, pp 40–69.
Schneider, M.R. A Multistage Process for Computing Virtual Dipolar Sources of EEG discharges from Surface Information, IEEE Tr. on Biomedical Engineering, Vol BME-19, No. 01, 1972, pp 1–12.
Schneider, M.R. Effect of Inhomogeneities on Surface Signals Coming from a Cerebral Current-Dipole Source, IEEE Tr. on Biomedical Engineering, January 1974. pp 52–54.
Smith D.B., Sidman R.D., Henke J.S., Flanigin H., Labiner D. and Evans C.N. Scalp and Depth Recordings of Induced Deep Cerebral Potentials, Electroenceph. and Clinical Neurophysiol, 1983, 55:145–150.
Stok C.J. The Influence of Model Parameters on EEG/EMG single Dipole Source Estimation, IEEE Tr. on Biomedical Engineering, Vol BME-34, 1987, pp 289–296.
Vaughan Jr.H.G. The Analysis of Scalp-Recorded Brain Potentials, in Bioelectric Recording Techniques, Part B: Encephalography and Human Brain Potentials, R.F. Thompson and M.M. patterson, Eds., Academic Press, New York, 1974, chapter 4, pp 157–207.
Watanabe Y., and Sakai Y. A graphic Method for Estimating Equivalent Dipole of Localized EEG Discharges, IEEE Tr. on Biomedical Engineering, Vol BME-31, no. 05, 1984, pp 435–439.
Weinberg H., Bricket P., Coolsma F., and Baff M. Magnetic localization of Intracranial Dipoles: Simulation with a Physical Model, Electroenceph. and Clin. Neurophysiology, 1986, 64: 159–170.
Witwer J.G., Trezek G.J., and Jewett D.L. The Effect of Media Inhomogeneities Upon Intracrania Electrical Fields, IEEE Tr. on Biomedical Engineering, Vol. BME-19, no. 05, 1972, pp 352–362.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Abeyratne, U.R., Kinouchi, Y., Oki, H. et al. Artificial neural networks for source localization in the human brain. Brain Topogr 4, 3–21 (1991). https://doi.org/10.1007/BF01129661
Accepted:
Issue Date:
DOI: https://doi.org/10.1007/BF01129661