Improved reconstruction of quantized CT scans via genetic algorithms
Chris Miller, Frank W. Moore, Brendan J. Babb, Michael R. Peterson · 2011
Cost-effective storage and timely transmission of medical images are very difficult technical challenges. Compression and reconstruction techniques must guarantee no significant loss of clinical information. This paper presents a convenient technique for improving the quality of reconstructed computed tomography (CT) images previously subjected to specified levels of lossy compression. Our genetic algorithm (GA) evolves novel transforms that consistently outperform state-of the-art wavelet-based schemes supported by the Digital Imaging and Communication in Medicine (DICOM) standard.