Binary medical image compression using the volumetric run-length approach

ABSTRACT

Image compression has become an inevitable tool along with the advancing medical data acquisition and telemedicine systems. The run-length encoding (RLE), one of the most effective and practical lossless compression techniques, is widely used in two-dimensional space with common scanning forms such as zigzag and linear. In this study, an algorithm which takes advantage of the potential simplicity of the run-length algorithm is devised in a volumetric approach for three-dimensional (3D) binary medical data. The proposed algorithm, namely 3D-RLE, being different from the two-dimensional approach that utilizes only intra-slice correlations, is designed to compress binary volumetric data by employing also the inter-slice correlation between the voxels. Furthermore, it is extended to several scanning forms such as Hilbert and perimeter to determine an optimal scanning procedure coherent with the morphology of the segmented organ in data. The algorithm is employed on four datasets for a comprehensive assessment. Numerical simulation results demonstrated that the performance of the algorithm is 1:30 better than those of the state-of-the-art techniques, on average.

GRAPHICAL ABSTRACT

KEYWORDS:

• 3D medical image compression
• lossless coding
• run-length encoding

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes on contributors

Erdoğan Aldemir received the B.Sc. degree from Uludağ University in 2010 and the M.Sc. degree from the Yıldız Technical University in 2013, all in electronics engineering. He is a Ph.D. candidate at Dokuz Eylül University and currently working on medical imaging, source coding, and image compression.

Gulay Tohumoglu was born in Gaziantep, Turkey. She received the B.S., M.S. and Ph.D. degrees in Electrical and Electronics Engineering from the Middle East Technical University METU, Ankara, Turkey, in 1983, 1986 and 1993, respectively. She was a visiting scientist at the Universidad Politechnica de Madrid, E.T.S.I de Telecommunicacion, Depart. de Senales, Sistemas Radiocommunicaciones-Spain from 1996 to 1997. She worked as Research assistant, instructor and assistant professor at the METU-Gaziantep Campus from 1983 to 1987, and as Associate Professor and Professor at the Gaziantep University from 1987 to 2010. She joined the Electrical and Electronics Engineering Dept. of Dokuz Eylül University, İzmir, Turkey where she is Professor at present. Her current research interests include signal and image processing, coding, wavelet transforms, biomedical systems and modelling of biological systems, deep learning, and analysis of circuit and systems.

M. Alper Selver received the B.Sc. degree from Gazi University, Ankara, Turkey, in 2002, and the M.Sc. and Ph.D. degrees from Dokuz Eylul University, Izmir, Turkey, in 2005 and 2010, respectively, all in electrical and electronics engineering. During his graduate studies, he has studied in Medical Informatics Laboratory at FH-Aachen Abt. Juelich, Germany and Heffner Biomedical Imaging Laboratory at Columbia University, New York, USA. Since 2011, he has been working as an assistant professor at the DEU EEE. His main research interests include the field of radiological image processing visualization, hierarchical, and multiscale classification strategies for biomedical applications and software development for the use of the developed techniques in clinic.

ORCID

Erdoğan Aldemir http://orcid.org/0000-0003-4772-8317

Gulay Tohumoglu http://orcid.org/0000-0003-2292-2148

Additional information

Funding

This work was supported by The Scientific and Technological Research Council of Turkey (TÜBİTAK) -Scientific and Technological Research Projects Funding Program- under grant number 116E133.