Abstract
Wireless electrocardiograms are useful for several practical applications in the healthcare domain. However, their usefulness is often limited by the quality of data that can be extracted from them. One of the main factors affecting the quality of ECG data is the inclusion of movement induced artefacts. In this paper we propose an adaptive filter to improve the quality of measurements. We propose to use motion or inertial sensors to capture the movements which affect the electrodes of a wireless ECG. Thus, we regard measurements from a 3D accelerometer and a 3D gyroscope as indication of the magnitude of noise artefacts in the outputs of a wireless ECG and use them to estimate and remove the movement artefacts. The paper presents the design and implementation of the filter, which we used to improve actual measurements we took whilst different subjects carried out various everyday activities (walking, running, riding a bicycle, and climbing up and down a staircase).