Abstract： In order to explore an electrocardiograph(ECG)compression method available for clinical diagnosis with high compres-sion ratio,the adaptive parameter optimal searcher was proposed to locate the parameters with good compression performance for ECG compression.A set of parameters applicable to all ECG signals was recommended for the universality of the algorithm,and the compres-sion performance was estimated by four indicators on the MIT-BIH database.The experimental results showed that the average compres-sion ratio(CR)was 26.67,the average percentage root-mean-square difference(PRD)was 14.64%,and the average compression time of 30 min ECG was 0.125 8 s.The results elucidate that the improved compression algorithm has outstanding performance in compres-sion ratio as well as application significance in clinical diagnosis.