Computational method for estimating boundary of abdominal subcutaneous fat for absolute electrical impedance tomography

Abdominal fat accumulation is considered an essential indicator of human health. Electrical impedance tomography has considerable potential for abdominal fat imaging because of the low specific conductivity of human body fat. In this paper, we propose a robust reconstruction method for high‐fidelity conductivity imaging by ion of the abdominal cross section using a relatively small number of parameters. Toward this end, we assume homogeneous conductivity in the abdominal subcutaneous fat area and characterize its geometrical shape by parameters defined as the ratio of the distance from the center to boundary of subcutaneous fat to the distance from the center to outer boundary in 64 equiangular directions. To estimate the shape parameters, the sensitivity of the noninvasively measured voltages with respect to the shape parameters is formulated for numerical optimization. Numerical simulations are conducted to demonstrate the validity of the proposed method. A 3‐dimensional finite element method is used to construct a computer model of the human abdomen. The inverse problems of shape parameters and conductivities are solved concurrently by iterative forward and inverse calculations. As a result, conductivity images are reconstructed with a small systemic error of less than 1% for the estimation of the subcutaneous fat area. A novel method is devised for estimating the boundary of the abdominal subcutaneous fat. The fidelity of the overall reconstructed image to the reference image is significantly improved. The results demonstrate the possibility of realization of an abdominal fat scanner as a low‐cost, radiation‐free medical device. To reconstruct an image robustly, we devised an estimation method to obtain the boundary of abdominal subcutaneous fat. The conductivity images with combination of patterns and sizes were reconstructed with a small systemic error of less than 1% in the estimation of the subcutaneous fat area. The fidelity of the reconstructed image is shown to significantly improve, and these results demonstrated the possibility of realization of a novel abdominal fat scanner that may contribute to the diagnosis of metabolic syndrome.