Noise is an important factor of the medical image quality, because the high noise of medical
imaging will not give us the useful information of the medical diagnosis. Basically, medical
diagnosis is based on normal or abnormal information provided diagnose conclusion. In this
paper, we proposed a denoising algorithm based on Contourlet transform for medical images.
Contourlet transform is an extension of the wavelet transform in two dimensions using the
multiscale and directional filter banks. The Contourlet transform has the advantages of multiscale
and time-frequency-localization properties of wavelets, but also provides a high degree of
directionality. For verifying the denoising performance of the Contourlet transform, two kinds of
noise are added into our samples; Gaussian noise and speckle noise. Soft thresholding value for
the Contourlet coefficients of noisy image is computed. Finally, the experimental results of
proposed algorithm are compared with the results of wavelet transform. We found that the
proposed algorithm has achieved acceptable results compared with those achieved by wavelet
transform.