Disclosed is a system for detecting pathophysiological cardiac conditions. The system comprises a diagnostic device that contains electronic circuitry that can detect a cardiac event such as an acute ischemia. The cardiac diagnostic device receives electrical signals from subcutaneous or body surface sensors. The cardiac diagnostic device includes a processor that computes QRS onset and offset points and fiducial points associated with T and U waves. The processor than baseline corrects the original signal/waveform by fitting a polynomial function to QRS offset points, and subtracting this function from the original waveform. Based on the baseline adjusted signal and/or the above mentioned fiducial points, the processor then computes averages of various waveform feature values, including a QRS measure sensitive to QRS curvature, T wave timing measures, ST segment deviation (difference between signal amplitudes at QRS offset and onset and/or minimum amplitude between QRS offset and peak T wave) and T/U wave amplitudes. These averages are computed by exponential averaging. From the exponential averages, the processor computes an average of the change in the averages over time. Based on the averages and the change in the averages, the processor applies an ischemia test to determine a likelihood of ischemia.
