Noise Reduction in Biomedical Images

Tomographic imaging modalities MRI,OCT & PET are§proving to be immensely helpful in the diagnosis of§various pathologies in clinical§environments. Images from these modalities span the§range from organ level to tissue level and represent§the condensation of information associated with the§objects that are imaged. The potential of this§information increases §signi cantly, if these images are effectively§processed and analyzed. However, the§quality of medical images is,usually, compromised due§to the undesirable noise§which reduces the SNR of these images. The degradation of§images by noise in uences the diagnosis of critical§diseases in clinical environments.§A novel noise reduction method based on wavelets and§differential evolution has§been proposed in this work. The proposed method was§tested on the images that§were acquired from MRI, OCT and PET. The application§of this method improved§SNR by a cosiderable amount while the loss in CNR was§kept at minimum level. A comprehensive analysis of§processed images, on the basis of different image§quality indices, showed that the proposed method,§effectively,§reduces noise without sacri cing the image resolution.