Diagnosis of cardiac pathology through prediction and approximation methods

The heart rate signal contains valuable information and its analysis has proven very useful in distinguishing healthy subject cardiograms from those of subjects with a variety of cardiac pathologies. The approach proposed here introduces a new use of prediction and approximation methods for the differentiation between healthy and unhealthy subjects. Local linear prediction and least squares approximation, though common in signal analysis, have yet to be applied as a cardiogram categorization tool. We show that measuring the error of both methods, as they are applied to each cardiogram, results in a clear distinction between the two groups. This is in coherence with cardiac physiology, since the behavior of a healthy subject ECG is more erratic than an unhealthy subject's.