BioImpedance Spectroscopy (BIS) have been investigated in many research areas as a method to detect changes in living tissues. However, BIS measurements are known to be hardly reproducible in clinical applications. This article proposes segmental spectral decomposition as a method of extracting reproducible parameters from raw BIS. The efficiency of this method is then compared to conventional Cole-Cole parameter extraction in a classification task.

International Steering Committee on Electrical Impedance Tomography
15th International Conference on Biomedical Applications of Electrical Impedance Tomography (EIT 2014)
International Steering Committee on Electrical Impedance Tomography

Nejadgholi, Isar, Batkin, Izmail, Bolic, Miodrag, Adler, A, & Shirmohammadi, Shervin. (2014). Segmental Spectral Decomposition as a Time Persistent Method of BioImpedance Spectroscopy Feature Extraction. In 15th International Conference on Biomedical Applications of Electrical Impedance Tomography (EIT 2014). International Steering Committee on Electrical Impedance Tomography.