Abstract
The widely used standard for medical image storage and transmission is named as Digital Imaging and Communication in Medicine (DICOM). In every field of medicine including diagnosis, treatment, and research, medical images that are obtained as the outputs of the techniques such as the Computerized Tomography (CT), magnetic resonance (MR), digital subtraction angiography (DSA) and Ultrasonography (US) are saved as DICOM format. Network sharing of these larger sized radiology images require large bandwidth. Hence before transferring, compression of such larger image files is necessary for easy and faster communication even with lower bandwidth. Huge amount of data either in multidimensional or multiresolution form is been created as a result of medical information. This makes the following steps like retrieval, efficient storage, management, and transmission of these data a complex process. This complexity could be reduced by compressing the medical data without any loss. Many methods have been proposed so far for compression of the large DICOM images, however with some limitations. Thus, specific methods to overcome the limitations like reducing the noise of MSE error signal and improving the PSNR value results in the medical images are to be proposed for the study. One such method is referred as Hybrid Weibull Probability Density Function based Continuous Wavelet based controulet transform (WPDF-CWBCT) that helps for compression of medical images without any data loss and also for improving the PSNR and reducing the MSE of the signal. The directional filter banks are being applied by initializing using the wavelet transform such that the image coding scheme is maintained based on the proposed transform. WPDF-CWBCT also uses a new set partitioning in hierarchical trees by employing a sorting method (SPIHT) algorithm that provided an embedded code. In this method, the diagnostics capabilities are not compromised to ensure the better performance of image compression is also been justified by the combination of wavelet based controulet transform and SPIHT. The performance evaluation of different DICOM and medical images is possible by using parameters like PSNR, MSE and image compression quality measures.
Highlights
The growing demand for high speed image transmission, efficient image storage and remote treatment can be met out by introducing an efficient image compression technique
Hybrid wavelet based controulet transformation for image compression: In contrast to the Laplacian pyramid used in contourlets, two stages of filter bank transformation results than normal Continuous Wavelet Transformation (CWT) results by changing the values of the image scale followed by calculation of its decomposition values using: x{I, I{ = #
The image samples are taken from Digital Imaging and Communication in Medicine (DICOM) sample image sets from http://www.osirix-viewer.com/datasets/. These image samples results are measured using the parameters like Peak Signal to Noise Ratio (PSNR), Mean Square Error (MSE), Root Mean Square Error (RMSE)
Summary
The growing demand for high speed image transmission, efficient image storage and remote treatment can be met out by introducing an efficient image compression technique. This study reviews about the different image compression techniques to arrive at a systematic approach for improving the performance of medical image compressing. Most of the electronic medical records constitute medical images as key components. The storing of these medical records along with its medical images requires enormous space. This makes the network sharing difficult as it consumes more time for transferring such medical images. These images comprising large set of data could be managed by DICOM standard as it offers reliable transfer of medical images when sharing patient’s record. On the whole the performance of PACS systems in storing the diagnostic images and other DICOM contents is been improved
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
