Silicon Implementation of the SPIHT Algorithm for Compression of ECG Records
Marc Simon Wegmueller, David Perels, Tobias Blaser, Stephan M. Senn, Philipp Stadelmann, Norbert Felber, Wolfgang Fichtner · Conference proceedings · 2006
Wavelet algorithms with subsequent encoders are used in the newest image compression standards such as JPEG2000 including EBCOT. The set partitioning in hierarchical trees (SPIHT) algorithm belongs to the next generation of encoders for wavelet-transformed images employing more sophisticated coding. SPIHT utilizes inherent redundancy among wavelet coefficients and is especially suited for electrocardiogram (ECG) data and color image compression. In this work, we present the first VLSI implementation using a modified SPIHT algorithm (MSPIHT). Compression ratios of up to 20:1 for ECG signals lead to acceptable results for visual inspection by medical doctors.