表面筋電図逆解析による活動電流源の推定

抄録

The in vivo analysis of bioelectric and biomagnetic potentials supplies valuable information for localization of current source. Only a few attempts, however, have so far been made at inverse analysis of surface EMG. The purpose of this study is to estimate functional and structural parameters of the motor unit (MU), e. g. depth and strength of the current source through the inverse analysis of surface EMG. Circumferential distribution of motor unit action potential (MUAP) on the skin surface was measured with a surface electrode array which was composed of 16 bipolar contacts. One-dimensional (circumferential direction) potential distribution at the peak time of MUAP and two-dimensional (circumferential direction and time) potential distribution were obtained by averaging of recorded surface EMG. The averaged potential was compared with the potential calculated by a model of current dipole propagating along the muscle fiber. In the model calculation, the depth, strength and position of the current source, conduction velocity (CV), and electrical conductivity were treated as unknown parameters and determined by minimizing the difference between the measured and the calculated potentials. The minimizing calculation was executed by Simplex method. The estimated depth of the current source was 1-6mm from the skin surface. Cross-sectional size of the motor unit estimated by the circumferential potential distribution was consistent with the results of measurement by multiple contact needle electrode and of histochemical experiment. Distribution in muscle fiber direction of the neuromuscular junctions belonging to single MU was estimated to range from 3mm to 15mm. If CV of MUAP and electrical conductivity of muscle are obtained by experiment or inverse analysis, we may estimate more accurately the depth, strength and size of the current source. Furthermore, in order to consider inhomogeneity and anisotropy of muscle tissue and boundary condition, we have to use numerical analysis tools such as finite element method.