Contour extraction of the left ventricular cavity from digital subtraction angiograms using a neural edge detector

摘要

In this paper, we propose a novel edge detector using a multilayer neural network, called the neural edge detector (NED), for extracting the contours which have good congeniality with those extracted by humans. A method for contour extraction from left ventriculographs with the lack of uniformity of contrast using the NED is presented. The proposed method consists of (a) edge detection by the NED, (b) extraction of rough contours, and (c) contour tracing. Experimental results have demonstrated that the proposed method can extract the contours which have good congeniality with those extracted by a medical doctor.