Denoising Weighted Bilateral Filter and Curvelet Transform Technique on MRI Wrist Joint Image

Abstract

One of the disturbances in image information on the magnetic resonance imaging (MRI) wrist joint was the emergence of noise due to the acquisition process of the use of radiofrequency (RF) coil instruments, so that a denoising technique was needed. Denoising weighted bilateral filter and curvelet transform (WBFCT) technique was applied to coronal image of MRI wrist joint T2 turbo spin echo (TSE) fat suppression (FS) spectral attenuated inversion recovery (SPAIR) sequences to analyze differences in image information between before and after denoising WBFCT on image of the sequence.
This study was a comparative analytical with a quasi-experimental pretest posttest without control group design. The study used 16 volunteers with qualitative image evaluation used 3 radiologists with statistical data analysis and quantitatively used mean square error (MSE) and peak signal to noise ratio (PSNR) values as a result of post processing images.
Information on the coronal MRI wrist joint image T2 TSE FS SPAIR sequence has different image information between before and after denoising WBFCT (p-value <0.001), with a mean of MSE 387,995 and PSNR 70,5369 decibel (dB), respectively, after a denoising image WBFCT had more optimal image information than before denoising WBFCT.
Denoising WBFCT can produce information on the coronal MRI wrist joint image T2 TSE FS SPAIR sequences optimally.