Building on the progress that has been made in the use of recurrent neural networks to solve optimisation problems, a new self-adaptive neural network is proposed that promises to improve upon existing image reconstruction techniques in Electrical Impedance Tomography. Interpreting image reconstruction in terms of interacting relaxations processes, a neural network formulation emerges that naturally leads to a parallel, cooperative algorithm that can easily be mapped to a massively parallel processor. The dynamics of the network are surprisingly simple and retain the possibility of analog implementation. Compared to algorithms currently used in EIT, the proposed method differs in its emphasis on relaxation processing and rigorous mathematical support.
Truyen, B & Cornelis, J 1993, Image reconstruction in Electrical Impedance Tomography: A selfadaptive neural network approach. in Proceedings 15th Annual International Conference IEEE Engineering in Medicine & Biology Society. vol. 1, IEEE Engineering in Medicine & Biology Society (EMBS), Piscataway, NJ, pp. 72-73,
15th Annual International Conference IEEE Engineering in Medicine & Biology Society, San Diego, California, United States, 28/10/93. https://doi.org/10.1109/IEMBS.1993.978398
Truyen, B., & Cornelis, J. (1993). Image reconstruction in Electrical Impedance Tomography: A selfadaptive neural network approach. In Proceedings 15th Annual International Conference IEEE Engineering in Medicine & Biology Society (Vol. 1, pp. 72-73). IEEE Engineering in Medicine & Biology Society (EMBS). https://doi.org/10.1109/IEMBS.1993.978398
@inbook{842a1635ffee40cc8035c84100beba75,
title = "Image reconstruction in Electrical Impedance Tomography: A selfadaptive neural network approach",
abstract = "Building on the progress that has been made in the use of recurrent neural networks to solve optimisation problems, a new self-adaptive neural network is proposed that promises to improve upon existing image reconstruction techniques in Electrical Impedance Tomography. Interpreting image reconstruction in terms of interacting relaxations processes, a neural network formulation emerges that naturally leads to a parallel, cooperative algorithm that can easily be mapped to a massively parallel processor. The dynamics of the network are surprisingly simple and retain the possibility of analog implementation. Compared to algorithms currently used in EIT, the proposed method differs in its emphasis on relaxation processing and rigorous mathematical support.",
keywords = "Electrical Impedance Tomography, neural networks, image reconstruction",
author = "Bart Truyen and Jan Cornelis",
year = "1993",
doi = "10.1109/IEMBS.1993.978398",
language = "English",
isbn = "978-0-7803-1377-4",
volume = "1",
pages = "72--73",
booktitle = "Proceedings 15th Annual International Conference IEEE Engineering in Medicine & Biology Society",
publisher = "IEEE Engineering in Medicine & Biology Society (EMBS)",
note = " 15th Annual International Conference IEEE Engineering in Medicine & Biology Society ; Conference date: 28-10-1993 Through 31-10-1993",
}