MRI video edge detection based on Fuzzy inference technique

Abstract

Objects recognition of the Magnetic Resonance Imaging (MRI) is a powerful tool in the detection of the geometric parameters such as area and perimeter. When outlining the border of an object, it is hard to calculate the area when it has irrelevant details inside the outline. Therefore, in this work we presented and applied a new technique based on fuzzy inference technique which helps to identify objects across MRI video with more accurate way. Fuzzy video edge detection technique was compared with Canny operator technique based on two assessment metrics (FOM and Pinho). Fuzzy video edge detection technique is superior than Canny operator with 14.5% and 7.5% based on the mean values of quantitative assessments (FOM and Pinho) respectively. The results are promising and it is possible to implement this technique and apply it on other medical imaging videos.