In this paper, a novel ECG data compression method is presented which employs set partitioning in hierarchical trees algorithm (SPIHT), sub-band energy compression (SEC) method and two-dimensional electrocardiogram (2D-ECG). The 2D-ECG is a two-dimensioned array, which each row of this array indicates one or more period and amplitude normalized (PANed) ECG beats. In the previous works, SPIHT and sub-band energy compression method have used to compress one or two-dimensional signals separately and have shown their efficiencies such as precise rate control, progressive quality, high compression ratio and low root mean square difference (PRD). In here, we put these two methods together and illustrate that they can be applied to 2D-ECG to achieve better results and this is because of the wavelet transform eliminating effect on redundancies between adjacent samples - it is also eliminated in one-dimensional ECG - and between adjacent beats by applying 2D wavelet transform