Medical Image Compression and Feature Extraction using Vector Quantization, Self-Organizing Maps and Quadtree Decomposition

Vector Quantization (VQ) is an efficient image compression approach. Among the different existing algorithms, Kohonen's Self Organizing Feature Map (SOFM) is one of the wellknown method for VQ. It allows efficient codebooks design with interesting topological properties to be performed. Furthermore, use of VQ for compression delivers basic information on the image content in the same process. However, in order to preserve the diagnostic accuracy in medical applications, the block size must be restricted to small values (e.g. 3x3, 4x4), which limits the compression rate. We propose to improve the compression performance by using several codebooks containing codewords of different sizes, according to the quadtree decomposition of the images. Results are compared to those provided by the standard JPEG image compression algorithm. Finally we introduce and discuss the signature maps of images using compression information.