Estimates of current density distributions: I. Applying the principle of cross-entropy minimization to electrographic recordings
- 1 December 1993
- journal article
- Published by IOP Publishing in Inverse Problems
- Vol. 9 (6) , 623-639
- https://doi.org/10.1088/0266-5611/9/6/002
Abstract
We consider the generalized bioelectromagnetic inverse problem (the evaluation of expressions for the unknown current densities that give rise to experimentally measurable field/surface potential readings). We apply the principle of cross-entropy minimization to compute new expressions for the expected means and convariances of the current densities. The new expressions are shown by computer simulations to offer better results than the existing methods in those cases when only partial information about the source is available. Our expressions are general enough to be equally valid for the biomagnetic, bioelectric or combined bioelectromagnetic inverse problems.Keywords
This publication has 25 references indexed in Scilit:
- Comparison of magnetoencephalography with other functional imaging techniquesClinical Physics and Physiological Measurement, 1991
- Continuous probabilistic solutions to the biomagnetic inverse problemInverse Problems, 1990
- Biomagnetic Fourier imaging (current density reconstruction)IEEE Transactions on Medical Imaging, 1990
- Probabilistic methods in a biomagnetic inverse problemInverse Problems, 1989
- The solution of the biomagnetic inverse problem by maximum statistical entropyInverse Problems, 1989
- Cortical magnetic fields evoked by frequency glides of a continuous toneElectroencephalography and Clinical Neurophysiology, 1982
- Image reconstruction from incomplete and noisy dataNature, 1978
- The Extracellular Potential Field of the Single Active Nerve Fiber in a Volume ConductorBiophysical Journal, 1968
- A Mathematical Evaluation of the Core Conductor ModelBiophysical Journal, 1966
- Information Theory and Statistical MechanicsPhysical Review B, 1957