This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.
Fig. 1
The FEM forward mesh based on an MRI image. White, red, and blue part indicate SAT layer, muscle, and bone, respectively.
Fig. 2
Experimental setup.
Fig. 3
Schematic diagram of the experimental schedule. After the 3rd measurement (t = 75 min), the subject did exercise to promote varying water contents compared to the original physiological state at the beginning of the experiment.
Fig. 4
(a) the normalized absolute conductivity distribution of the right lower leg (viewed from the top) reconstructed by absolute EIT at 1st measurement; These are used as inhomogeneous reference data in (b). Time series of reconstructed relative conductivity distribution using the image-based reference EIT (b) and conventional time difference EIT (c) at 2nd, 3rd, and 4th measurement, respectively.
Fig. 5
The normalized temporal variation over segmental conductivity σseg in the predicted subcutaneous layer (white part in Fig.1) and segmental extracellular water volume Vseg in the right leg from image-based reference EIT (IBR-EIT) and MFBIA data, respectively.
Fig. 6
The magnitude of impedance at 0 min and 60 min on Day 1.
Schematic flow of image-based reference EIT.
Step 1:
Defining the filtering method to eliminate the muscle’s conductivity distribution and unexpected noise background
Obtain:Equation (4)
Step 2:
Using voltage data t = 0 as an initial conductivity distribution, selecting the mesh of SAT on the forward problem mesh condition, calculating the fat weighted threshold value\overline {{{\hat \sigma}_{fat}}} + std\left({{{\hat \sigma}_{fat}}} \right)