Image processing algorithms for medical applications

Abstract

Image processing algorithm is used to reduce the noise present in medical images. This is a crucial procedure as medical images are usually corrupted by noise during transmission, acquisition or processing. The presence of noise results in distortion of the image, affecting the image quality, thus causing the inaccuracy of image analysis by the doctors and may lead to misdiagnosis. Noise reduction is vital for doctors to diagnose the patients’ illness accurately and quickly, proposing a suitable treatment for the patients. In this paper, noise removal will be done using these four filters (i.e. Gaussian, Mean, Median and Wiener filter). The performance evaluation parameters used to measure the effectiveness of each filter are Signal to Noise Ratio (SNR), Mean Squared Error (MSE), Peak Signal to Noise Ratio (PSNR) and Structural Similarity Index Measure (SSIM). The four types of medical images used in noise removal are Magnetic Resonance Imaging (MRI), Computed Tomography (CT), X-rays and Ultrasound. The four types of medical images will be corrupted individually with four different noises (i.e. Gaussian, Poisson, Salt and Pepper and Speckle noise). The experiment was carried out on a total of 40 images, 10 images for each type of medical images. By comparing the averaged performance of each filter on the four different types of medical images that are corrupted by the four types of noises, conclusions of the effectiveness of the filters can be made.