A differential evolution approach to noise reduction in biomedical images

Abstract

Tomographic imaging modalities such as Magnetic Resonance Imaging (MRI), Optical Coherence Tomography (OCT) and Positron emission tomography (PET) are proving to be immensely helpful in the diagnosis of various pathologies in clinical environments. Images obtained using these modalities span the range from organ level to tissue level and represent the condensation of the information associated with the objects that are imaged. The potential of this information increases significantly, 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 signal to noise ratio (SNR) of these images. The degradation of images by noise influences 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 contrast to noise ratio (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 sacrificing the image resolution.