A number of composers have explored applications of complexity ("chaos" theory) to music composition and synthesis [1]. Heart rhythms are also not new to musical contexts [2,3], and experimentas have even been conducted mapping heart rhythms in real time to brief auditory displays for use as biofeedback [4]. The impetus of this project, however, takes a different focus. Rather than setting out to create musically interesting sounds, or real-time monitoring, the approach here is to explore whether physiological variations in heart rate dynamics over a period of hours could be a source of medically useful sounds. In other words, can bedside diagnosis be aided by information taken from an auditory display of heart rate variability?

Heart rate fluctuations can be readily measured from the electrocardiogram (ECG), a graphical recording of the electrical potentials generated by cardiac muscle cells. While clinicians often refer to the healthy heartbeat as "regular sinus rhythm", healthy subjects typically display more complex patterns than those found in unhealthy ones [5-7]. For example, following a heart attack, patients whose heart rates are overly regular may be at increased risk of fatal cardiac arrhythmias [8].

Electrocardiographic recording of the heartbeat.
The QRS complexes represent electrical activation of the ventricles.
The RR interval is the time between consecutive QRS complexes.

To measure heart rate over extended periods, physicians make use of Holter monitors, ECG devices that permit long-term ambulatory recording and storage of ECG waveforms for periods on the order of 24 hours. Following the recording, the data can be processed via automated or semi-automated programs. Such programs detect the electrical pulses, termed QRS complexes, that trigger mechanical contraction of the heart's pumping chambers (ventricles). Further analysis of these QRS waveforms can be used to generate a sequence of intervals (the intervals between QRS complexes, also called NN intervals) that represent the time periods between consecutive normal beats. These RR intervals, representing heart rate fluctuations, are the data sets most often used in heart rate variability analysis.

The heart's normal beats are initiated by impulses from pacemaker cells in the sinus node, hence the term normal sinus rhythm. The sinus node frequency is modulated primarily by input from the autonomic (involuntary) nervous system. There are two major components of this system: the sympathetic -- which increases heart rate -- and the parasympathetic (or vagal) -- which decreases the heart rate. The nonlinear interaction between these two competing components is responsible for much of the heart rate's complex intrinsic fluctuations. Mechanical and other metabolic influences may also contribute to HRV. A major factor regulating heart rate variations over the short term are the effects of respiration which are mediated via the parasympathetic branch of the autonomic nervous system. During inspiration heart rate typically increases, while during expiration it decreases. These oscillations are referred to as respiratory sinus arrhythmia.  Pathologic breathing patterns such as those seen with obstructive sleep apnea may be associated with lower frequency oscillations in heart rate [9, 10].

Diverse measures of heart rate variability have been proposed and some of these may be associated with a variety of cardiopulmonary and systemic disorders [11, 12]. However, implementation of such methods remains difficult to interpret on an individual basis, and in the absence of a consensus as to their utility, these measures still have limited practical bedside applicability at present.

The auditory system is particularly well suited for following multiple streams of information[13-15] such as those contained in complex heart beat time series. With this in mind, sonification may offer an effective means for simultaneous display of many signal processing operations. By displaying an HRV data set as a multidimensional sonification, correlations among analytic techniques, which might normally be difficult to perceive, may be observed aurally. Further, such a technique might prove useful in screening long time series records for clinically important dynamics.