ENHANCEMENT TECHNIQUES FOR ABDOMINAL ORGANS’ MAGNETIC RESONANCE IMAGING

Abstract

Medical images which carry important information about the human body most times appear with
low visual quality; hence, it is often strenuous to detect and extract adequate information from them.
Information extracted can lead to right or wrong diagnosis and prognosis. To obtain optimum
images for accurate diagnosis therefore, these images must pass through an enhancement process
which consists of a series of methods that aim at improving their visual assessment. Magnetic
Resonance Imaging (MRI) produces most of the important medical images of which the raw data
acquired can be corrupted by several types of noise and artefacts. This paper aims at improving the
quality of abdominal MR images in the spatial domain. The method began with the application of
the median filter for the noise reduction. Thereafter, the filtered image was further enhanced with
Histogram Equalization (HE), Contrast Limited Adaptive Histogram Equalization (CLAHE), and a
proposed technique; the Power Law Transformed Adaptive Histogram Equalization (PLTAHE). A
sharpening effect was applied using the unsharp mask filter for the result. The results of the applied
techniques were compared side by side for the best enhancement effect, and an objective assessment
was carried out on by evaluating the Peak Signal to Noise Ratio (PSNR) and Structural Similarity
Index Measure (SSIM) values of the images. A subjective assessment was also carried out with a
radiologist who gave a better perception to the effects of the enhancement techniques on the samples
studied. The results of the study showed that the proposed technique Power Law Transformed
Adaptive Histogram Equalization produced the best contrast for the analysed MR images.