12.12. New computational methods for electrical tomography imaging

Public examination of a doctoral dissertation in the field of Applied Physics
Doctoral candidate: M.Eng. Dong Liu
Date and venue: 12.12.2015 at 12 noon, SN201, Snellmania, Kuopio campus
Language of the dissertation and the public examination: English

Electrical impedance tomography (EIT) is an imaging modality in which the distribution of electrical conductivity inside a body is reconstructed based on measurements from the surface of the body. EIT has been applied to, e.g., imaging of industrial processes, non-destructive material testing and geophysical exploration. EIT has also potential for becoming a tool of medical imaging – especially for monitoring physiological processes, such as ventilation and heart function.

Reconstructing the conductivity distribution based on EIT measurements is an ill-posed inverse problem and thus highly intolerant to modeling errors, such as inaccuracy of the body shape. Mainly for this reason, many practical applications of EIT in medical imaging are based on linearized difference imaging in which the temporal change of the conductivity distribution is reconstructed based on two sets of measurements: before and after the change. Linearized difference imaging is relatively tolerant to modeling errors, but usually provides only qualitative information on the change of the conductivity.

In his PhD thesis work, Dong Liu, M.Eng, has developed a novel non-linear difference imaging approach to EIT. The results of his numerical and experimental studies of two- and three-dimensional (2D & 3D) EIT demonstrate that non-linear difference imaging tolerates modeling errors at least to same extent as linear difference imaging. Moreover, unlike the linear method, the non-linear difference imaging provides quantitative information on the conductivity change. The new reconstruction method is particularly advantageous in cases where the electrical conductivity is known to change only in a certain region in a body – the non-linear difference reconstructions with region-of-interest constraints outperform the conventional reconstructions in all studies of the thesis work.

The doctoral dissertation of M.Eng. Dong Liu entitled Non-linear Difference Imaging Approach to Electrical Impedance Tomography will be examined at the Faculty of Science and Forestry. The opponent in the public examination will be Assistant Professor Mohammad Pour-Ghaz, North Carolina State University and the custos will be Docent Aku Seppänen, University of Eastern Finland.