Reach Of ECG Compression Method And The Application On DSP System

Electrocardiogram diagnosis is developing towards remote and intelligent modewith the progress of technology. Data compression technology is needed for theefficient storage, analysis and transportation of the large amount of ECG data. And inthe case of remote patient monitoring, real-time ECG compression and transportationis required to improve the diagnostic efficiency and to provide medical treatment intime. This thesis is composed of an ECG compression algorithm based on waveletanalysis and its real-time application on DSP system. This multiple compression algorithm ,which is proposed by studying existingalgorithm, is suitable for real-time application. It makes use of the energyconcentration characteristic of wavelet coefficients and the relationship of waveletcoefficients in different scales to form quantization vectors. A codebookreplenishment mechanism is used in the process of vector quantization to decrease thecomputational complexity. An energy standard is set to reduce the data producedduring the codebook replenishment. The reconstruction distortion is controlledautomatically by a threshold adjustment mechanism to enhance the algorithm'srobustness . Finally, the quantization result is recompressed by lossless encoder. Thesimulation results indicated that the compression method is effective. The DSP system,using TMS320VC33 chip of TI as the kernel, is designed tocompress ECG signals in real time with the presented method. The hardware designincludes A/D channel design, external memory expansion design and serial interfacedesign. Sampling, compression and transportation of ECG signals are accomplishedby embedded system programming.The result of testing with real ECG data indicatedthis system could accomplish real-time compression of 3 leads ECG data with thesame performance of simulation.