Efficient compression of volumetric medical images using Legendre moments and differential evolution

Abstract

Volumetric medical images are widely used in diagnosing and detecting health problems of patients. Large datasets of volumetric medical images required huge storage space and high network capabilities to transmit these medical images from one location to another especially in the applications of telemedicine and teleradiology. In addition, the quality of medical images plays an important role in successful diagnoses. Therefore, an efficient compression algorithm must achieve significant reduction in the size of these volumetric medical images by using high compression ratio and preserve the quality of these images for successful diagnosis. In this paper, a novel optimized compression algorithm for volumetric medical images is proposed. In this algorithm, the volumetric medical images are divided into two-dimensional (2D) slices where each slice is divided into a group of 8 × 8 nonoverlapped blocks. The Legendre moments are computed for each block where the differential evolution optimization algorithm is utilized to select the optimum moments according to minimization of the cost function. Volumetric medical images from different medical imaging modalities are used in testing and evaluating the proposed compression algorithm. The performance of the proposed algorithm is compared with the existing volumetric medical images compression algorithms where the comparison clearly shows that the proposed algorithm outperforms the existing compression algorithms in terms of mean square error, peak signal-to-noise ratio, normalized correlation coefficient, and structural similarity index.