A method comprising sensing at least one cardiac signal representative of cardiac activity of a subject using an implantable medical device (IMD), calculating, from the cardiac signal, a first dominant vector corresponding to a direction and magnitude of maximum signal power of an ST-T first segment of a cardiac cycle and a second dominant vector corresponding to a direction and magnitude of maximum signal power of a P-QRS second segment of a cardiac cycle, measuring a change in the first dominant vector, measuring a change in the second dominant vector, and subtracting the change in the second dominant vector from the measured change in the first dominant vector to form a difference.