Data compression of electrocardiograms (ECG) has become increasingly important because of the proliferation of computerized ECG processing systems. Data compression is useful for storage and transmission of digitized ECGs. In this paper, we describe a new ECG data compression algorithm called SAIES. It combines the advantages of the AZTEC and Fan compression techniques. We evaluated and compared the SAIES performance to that of the AZTEC and Fan algorithms, utilizing an independent database under identical test conditions. A 90% improvement in compression ratio over the Fan and more than 50% over the AZTEC are achieved. The signal fidelity of SAIES outperforms that of the AZTEC algorithm. This is based on visual inspection and computation of the rms error between the original and the reconstructed ECG signal.