Least Mean Fourth Algorithm for Seismocardiography Signal Detections

During cardiac quiescence periods, the seismocar- diogram (SCG) can accurately stimulate the cardiac computed tomography angiography. From the SCG chest wall acceleration, it requires to compute the velocity as it is unavailable and corrupted by noise. This paper proposes a least mean fourth algorithm to compute the SCG states from the noisy SCG and measurement signals. Based on the steepest descent criteria, the proposed algorithm is developed. In order to improve system performance and novelty, the error function is defined by fourth power, which can reduce estimation errors lead to an accurate estimation. After minimization the objective function, the suboptimal algorithm is proposed for removing noise from the SCG and measurement signals. Numerical results show that the proposed method can accurate estimate the SCG states.