Document Type

Theses, Ph.D

Rights

This item is available under a Creative Commons License for non-commercial use only

Disciplines

2. ENGINEERING AND TECHNOLOGY, Electrical and electronic engineering

Publication Details

A thesis submitted to the Dublin Institute of Technology for the degree of Doctor of Philosophy (Ph.D.). August 2013.

Abstract

Certain areas of the body, such as the hands, eyes and organs of speech production, provide high-bandwidth information channels from the conscious mind to the outside world. The objective of this research was to develop an innovative wearable sensor device that records signals from these areas more conveniently than has previously been possible, so that they can be harnessed for communication. A novel bioelectrical and biomechanical sensing device, the wearable endogenous biosignal sensor (WEBS), was developed and tested in various communication and clinical measurement applications.

One ground-breaking feature of the WEBS system is that it digitises biopotentials almost at the point of measurement. Its electrode connects directly to a high-resolution analog-to-digital converter. A second major advance is that, unlike previous active biopotential electrodes, the WEBS electrode connects to a shared data bus, allowing a large or small number of them to work together with relatively few physical interconnections. Another unique feature is its ability to switch dynamically between recording and signal source modes. An accelerometer within the device captures real-time information about its physical movement, not only facilitating the measurement of biomechanical signals of interest, but also allowing motion artefacts in the bioelectrical signal to be detected. Each of these innovative features has potentially far-reaching implications in biopotential measurement, both in clinical recording and in other applications.

Weighing under 0.45 g and being remarkably low-cost, the WEBS is ideally suited for integration into disposable electrodes. Several such devices can be combined to form an inexpensive digital body sensor network, with shorter set-up time than conventional equipment, more flexible topology, and fewer physical interconnections.

One phase of this study evaluated areas of the body as communication channels. The throat was selected for detailed study since it yields a range of voluntarily controllable signals, including laryngeal vibrations and gross movements associated with vocal tract articulation. A WEBS device recorded these signals and several novel methods of human-to-machine communication were demonstrated. To evaluate the performance of the WEBS system, recordings were validated against a high-end biopotential recording system for a number of biopotential signal types. To demonstrate an application for use by a clinician, the WEBS system was used to record 12‑lead electrocardiogram with augmented mechanical movement information.

DOI

10.21427/D79903