An electrocardiogram (ECG) records the electrical activity of the heart using two electrodes to detect electrophysiological signals from the organ and one reference electrode (typically attached to the right leg) as the reference potential. An inconvenience of the three electrodes is that the grounds of the amplifiers are interconnected all over the body. To address this inconvenience, this paper proposes a two-electrode ECG measurement method with common electrodes instead of the reference electrode of the right leg. The power supply and the ground of the amplifier supplying the two electrode pairs are not connected and independent. Pasted on the skin with the body as the reference potential, both electrodes detect electrophysiological signals, transmitting the signals through two isolation amplifiers to an instrumentation amplifier filter and generating ECG waveforms. Both previous-stage electrode amplifier circuits of the isolation amplifiers used their respective power supplies; their grounds were unconnected to those of the power supplies. Similarly, the next-stage electrode amplifiers used their respective power supplies, with their grounds unconnected to those of the power supplies. Experiments yielded ECG signal waveforms in Leads I to III, with that of Lead I most resembling the waveforms of conventional reference electrodes. This method can be used to develop wireless ECG devices and portable ECG devices that eliminate the need for wiring to facilitate ECG measurement.