Techniques for event and feature detection

Event and feature detection are important aspects of signal processing in cardiology. Knowledge of the rate and rhythm of the heart are critical aspects of patient monitoring. Even within a single beat, the different phases of the cardiac cycle provide important information of the physiology. Although this information could be gleaned manually from the electrocardiogram, continuous blood pressure, or respiratory signals, having automatic calculations of the rates, rhythms, and amplitudes of these different signals nearly instantaneously saves precious time that could be used to save a patient's life. These real-time measurements are often also linked to visual or audio alarms that alert hospital staff when something is going wrong. Cardiac gating for image acquisition requires accurate identification of electrocardiographic features. Automatic event and feature detection is also useful in off-line analysis and in automated analysis for cardiac imaging (i.e., determining the heart's boundaries). This chapter will discuss common signal processing techniques that are used for event and feature detection.