Preface

Many areas of clinical practice and research share a common need for visualization and analysis of physiologic signals. Typically, signals such as electrocardiograms (ECG), respiration, blood pressure, electroencephalograms (EEG), electrooculograms (EOG), and electromyograms (EMG) may be acquired during a clinical procedure or experiment, for durations ranging from a few seconds to many hours. Often these signals must be monitored and analyzed in real time (i.e., while they are being acquired). In other cases, the signals can be recorded for later analysis.

There are two common reasons for recording signals for later analysis. First, the analysis may be too complex to perform in real time, or it may require observation of long time periods (for example, in Fourier spectral analysis of very low frequencies). Second, the analytic technique itself, as well as the signals, may be a subject of investigation. Digitally recorded signals are ideally suited as test material in such cases, since they can be used to provide strictly reproducible inputs to a variety of analysis methods.

WAVE is a computer program that helps you, the clinician or researcher, to analyze digitally recorded signals. Using WAVE , you can view any desired portion of your signals as if you were browsing through a chart recording. (WAVE can print a paper `chart recording' of any portion of the signals, if you wish.) You can annotate (label) any features of the signals you choose. You can select any subset of the signals, and any time interval, to be analyzed by an external program under the control of WAVE . If the analysis program generates signal annotations, you can view and correct them. Starting with example programs and library functions provided in the WFDB Software Package, you can write your own programs for signal analysis, and add their capabilities to WAVE 's repertoire, without recompiling (or even restarting) WAVE .

This guide describes how to use WAVE , how to extend its capabilities, and what hardware and software you will need in order to do so. It does not presume extensive familiarity with computers or signal processing, but it would be helpful to have a friend with some computer experience available when you begin.

Many friends have suggested improvements in WAVE . Thanks to Penny Ford-Carleton, Ted Clancy, Kevin Clark, Leon Glass, Scott Greenwald, Farzin Guilak, Jeff Hausdorff, Yuhei Ichimaru, David Israel, Franc Jager, Joe Mietus, Rama Mukkamala, Sheila Ryan, Kambiz Soroushian, Alessandro Taddei, and Andy Wieckiewicz, and to all of the early users of WAVE . I would especially like to thank Roger Mark for his continuous support and encouragement of this project.

Your comments on this guide, and on WAVE , are welcome. Please send them to:

PhysioNet
MIT Room E25-505A
Cambridge, MA 02139
USA

(e-mail: wfdb@physionet.org)