A Novel and Efficient Method for Denoising and
Compression of MR Images for Telemedicine
Applications
Journal:
GRENZE International Journal of Engineering and Technology
Authors:
Ananda Resmi S, Ajith A
Volume:
1
Issue:
1
Grenze ID:
01.GIJET.1.1.13
Pages:
53-58
Abstract
The presence of Rician noise is the major drawback of MR images for data
storage, transmission and analysis for medical applications especially for diagnostic
applications at remote locations. This work demonstrates an efficient method for removing
the Rician noise and compression of MR image for data storage and transmission
applications by preserving useful details. The method demonstrates a fast algorithm based
on non local means and DCT based compression scheme for effective storage and
transmission applications. This method was tested for 20 sets of MRI data for different noise
level and its Peak signal to noise ratio (PSNR) value found acceptable for further
applications. This algorithm was effectively degraded the presence of Rician noise which
reduced the number of computations and thus by minimising the computational complexity
and time of the system. The performance of the method is evaluated and compared by
means of mean square error and Peak Signal to Noise Ratio of existing algorithm and it is
observed that these parameters were outperformed with respect to existing algorithm
which is acceptable for further analysis of MRI Image.