Abstract
Purpose
This study was aimed to validate the performance of a novel image compression method using a neural network to achieve a lossless compression. The encoding consists of the following blocks: a prediction block; a residual data calculation block; a transformation and quantization block; an organization and modification block; and an entropy encoding block. The predicted image is divided into four macro-blocks using the original image for teaching; and then redivided into sixteen sub-blocks. The predicted image is compared to the original image to create the residual image. The spatial and frequency data of the residual image are compared and transformed.
Materials and methods
Chest radiography, computed tomography (CT), magnetic resonance imaging, positron emission tomography, radioisotope mammography, ultrasonography, and digital subtraction angiography images were compressed using the AIC lossless compression method; and the compression rates were calculated.
Results
The compression rates were around 15:1 for chest radiography and mammography, 12:1 for CT, and around 6:1 for other images. This method thus enables greater lossless compression than the conventional methods.
Conclusion
This novel method should improve the efficiency of handling of the increasing volume of medical imaging data.
Similar content being viewed by others
References
Lee KH, Lee HJ, Kim JH, Kang HS, Lee KW, Hong H, et al. Managing the CT data explosion: initial experiences of archiving volumetric datasets in a mini-PACS. J Digit Imaging 2005;18:188–195.
Kohs GJ, Legunn J. CT in your clinical practice. AJR Am J Roentgenol 2004;183(suppl):1–12.
Rubin GD. Data explosion: the challenge of multidetector-row CT. Eur J Radiol 2000;36:74–80.
Neelamani R, de Queiroz R, Fan Z, Dash S, Baraniuk RG. JPEG compression history estimation for color images. IEEE Trans Image Process 2006;15:1365–1378.
Wu D, Tan DM, Baird M, DeCampo J, White C, Wu HR. Perceptually lossless medical image coding. IEEE Trans Med Imaging 2006;25:335–344.
Reiter E. Wavelet compression of medical imagery. Telemed J 1996;2:131–137.
Zhang N, Wu X. Lossless compression of color mosaic images. IEEE Trans Image Process 2006;15:1379–1388.
Electronic Information Committee of The Japan Radiological Society. Digital image handling guidelines V2.0. Electronic Information Committee of The Japan Radiological Society; April 2006.
Azpiroz Leehan J, Lerallut JF, Magana I. A multiprocessor architecture for medical image compression in a PACS environment. Med Prog Technol 1994;20:101–110.
Fausett LV. Fundamentals of neural networks: architectures, algorithms and applications. Englewood Cliffs, NJ: Prentice Hall; 1994.
Rummelhart DE, McClelland JL, PDP Research Group. Parallel distributed processing. Cambridge, MA: MIT Press; 1986.
Ikeda M, Ito S, Ishigaki T, Yamauchi K. Evaluation of a neural network classifier for pancreatic masses based on CT findings. Comput Med Imaging Graphics 1997;21:175–183.
Tourassi GD, Floyd CE, Sostman HD, Coleman RE. Acute pulmonary embolism: artificial neural network approach for diagnosis. Radiology 1993;189:555–558.
Scott R. Artificial intelligence: its use in medical diagnosis. J Nucl Med 1993;34:510–514.
Baker JA, Kornguth PJ, Lo JY, Williford ME, Floyd CE Jr. Breast cancer: prediction with artificial neural network based on BI-RADS standardized lexicon. Radiology 1995;196:817–822.
Lo SC, Li H, Freedman MT. Optimization of wavelet decomposition for image compression and feature preservation. IEEE Trans Med Imaging 2003;9:1141–1151.
Hosaka T, Kabashima Y, Nishimori H. Statistical mechanics of lossy data compression using a nonmonotonic perception. Phys Rev E Stat Nonlin Soft Matter Phys 66(Pt 2):066126. Epub 2002 Dec 18.
Dony RD, Coblentz CL, Nabmias C, Haykin S. Compression of digital chest radiographs with a mixture of principal components neural network: evaluation of performance. Radiographics 1996;16:1481–1488.
Author information
Authors and Affiliations
Corresponding author
About this article
Cite this article
Fukatsu, H., Naganawa, S. & Yumura, S. Development and evaluation of a novel lossless image compression method (AIC: artificial intelligence compression method) using neural networks as artificial intelligence. Radiat Med 26, 120–128 (2008). https://doi.org/10.1007/s11604-007-0205-8
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11604-007-0205-8