Error Resilient Transmission and Security filtering of Medical Images

Abstract

Introduction Medical images should be subjected to loss-less compression, a technique that stems from mathematical theory of communication (Shannon, 1948). Loss-less compression techniques use variable length codes, (David Huffman, 1952). Compression ratio achieved is not very satisfactory in Huffman. Hence Run length encoding, yields a more effective compression algorithm that increases the compression ratio on medical images. Objectives - To provide lossless compression of Medical images by applying Wavelet transform and run length encoding - To provide security mechanism by eliminating the textual content in Medical images Proposed System With the DICOM standard, it is easy to eliminate textual information such as patient name and ID.We use Daubechies' wavelets and analysis techniques to detect the high frequency variation in the diagonal direction that is indicative of text. Only sensitive patient identification information is eliminated while retaining the medical information in the image. Encoding and decoding can be done by applying Run length technique. Excellent results have been obtained in experiments using a large set of real world medical images many with superimposed text. Methodology Matlab Software Version 7.0.1 consists of various modules: 1.The Input Module to retrieve the Medical Image as input. 2.Provide security feature by changing the DICOM unique identifier (UID). 3.Wavelet decomposition module to provide wavelet compression using Daubechies wavelet of order 2. 4.Compression Module to compress the input image by applying Run Length encoding. 5. Reconstruct original image from compressed image data applying Run Length decoding . 6.Wavelet reconstruction to decompress the image and extract the Original Image. Simulation Result and discussion Simulation: The Images used in this project are shown in the Figure below.The Images for transformation are scanned directly from IPRO GE SYTEC 1800-i CT SCANNER.These Images are in DICOM format and are then converted to .dcm. Results In this thesis we have developed a technique for wavelet transforms.Wavelet transform making it attractive both in terms of speed and memory needs and enhancing security features also. It is found that the proposed method gives more than 34% average improvement in the PSNR value in the bpp range of 0.0625 to 1.00 and highly reduction in Mean square error with a better quality of the reconstructed medical image judged on the basis of the human visual system (HVS). So, finally we can conclude that the proposed Wavelet based method is very suitable for low bit rate compression, high compression ratios, can perform lossless coding, high PSNR, low MSEs as well as good visual quality of the reconstructed medical image at low bit rates. It can also maintain the high diagnostic quality of the compressed image.