Abstract:
"
This paper analyzes the process of edge detection applied to liver sonograms and its
benefit in the medical domain for aiding in the analysis and diagnosis of the human
liver. To provide a solution, multiple edge detection solutions are compared with
each other as well as a proposed method in order to find the algorithm that works
best for this use case, which is for sonograms, after which evaluations are carried out. The results of this study shows that the edge detection of liver sonograms is
beneficial to sonographers, and that the proposed method for detecting edges in liver
sonograms is of expected quality and works better than previous methods in terms of
output quality, as well as speed. This study shows that the methods of edge detection
can be improved in quality and that the medical domain is in need of tools such as
the one implemented in this study. There is room for future work to improve the
proposed solution further and to implement more features that will be beneficial to
sonographers."
