Medical Image Compression Based on Region of Interest

Authors

1 Radiation Engineering Department, National Centre for Radiation Research and Technology (NCRRT), Egyptian Atomic Energy Authority, Cairo, Egypt

2 Nano electronics Integrated System Center (NISC), Nile University, Cairo, Egypt

3 Department of Information Technology, Cairo University, Cairo, Egypt

Abstract

Medical images show a great interest since it is needed in various medical applications. In order to decrease the size of medical images which are needed to be transmitted in a faster way; Region of Interest (ROI) and hybrid lossless compression techniques are applied on medical images to be compressed without losing important data. In this paper, a proposed model will be presented and assessed based on size of the image, the Peak Signal to Noise Ratio (PSNR),and the time that is required to compress and reconstruct the original image.
The major objective of the proposed model is to minimize the size of image and the transmission time. Moreover, improving the PSNR is a critical challenge.The results of the proposed model illustrate that applying hybrid lossless
techniques on the ROI of medical images reduces size by 39% and gives better results in terms of the compression ratio and PSNR.

[1] Yee D., Soltaninejad S., Hazarika D., Mbuyi G., Barnwal R., and Basu. Medical image compression based on region of interest using better portable graphics (bpg). In IEEE International Conference on Systems, Man, and Cybernetics (SMC 2017), pages 216–221, 2017.
[2] Tomar R.R.S. and Jain K. Lossless image compression using differential pulse code modulation and its application. In The 2015 Fifth International Conference on Communication Systems and Network Technologies), pages 543–545, 2015.
[3] S. Dayal and N. Gupta. Region of interest based compression of medical image using discrete wavelet transform. International Journal on Computational Science and Applications(IJCSA), 5:81–91, 2015.
[4] Joshua T.P., Arrivukannamma M., and Sathiaseelan J.G.R. Comparison of dct and dwt image compression. International Journal of Computer Science and Mobile Computing), 5:62–67, 2016.
[5] Swathy S. nad Jumana N. A study on medical image compression techniques. International Journal of Innovative Research in Computer and Communication Engineering, 5:8106–8110, 2017.
[6] Hasan T. Image compression using discrete wavelet transform and discrete cosine transform. Journal of Applied Sciences Researches, 13:1–8,2017.
[7] S. Kazeminia, Nader K., Reza S., Shadrokh S.,Harm Derksen, and Kayvan Najarian. Region of interest extraction for lossless compression of bone x-ray images. In International Conference of the IEEE Engineering in Medicine and Biology
Society (EMBS 2015), pages 3061–3064, 2015.
[8] Reddy B. V., Reddy P. B., P. S. Kumar, and Reddy A. S. Region of interest extraction for lossless compression of bone x-ray images. In IEEE 6th International Advanced Computing Conference (IACC 2016), pages 404–408, 2016.
[9] Lakshminarayana. M and Sarvagya M. Rm2ic: Performance analysis of region based mixedmode medical image compression. International Journal of Image Graphics and Signal Processing,9:12–21, 2017.
[10] B. Mohamed and H. Afify. Mammogram compression techniques using haar wavelet and quadtree decomposition-based image enhancement.Biomedical Engineering: Applications, Basis and Communications, 29:1750038(1)–1750038(7), 2017.

[11] Suckling J., Parker J., Dance D., Astley S.,Hutt I., Boggis C., Ricketts I., Stamatakis E.,Cerneaz N., Kok S., and Taylor P. The mammographic image analysis society digital mammogram database. In Exerpta Medica International
Congress Series, 1069:375–378, 1994.
[12] Manju M., Akila U. Abarna P., and Yamini S.Peak signal to noise ratio and mean square error calculation for various images using the lossless image compression in ccsds algorithm. International Journal of Pure and Applied Mathematics,
119:14471–14477, 2018.