2020 IEEE Nuclear Science Symposium and Medical Imaging Conference (NSS/MIC)(2020)
Faculty of Biology and Medicine
被引用0|浏览0
摘要
In this work, we present a whole-body image segmentation algorithm that is completely unsupervised. It uses diverse topological features of intensity distributions over different organs to perform segmentations. We compared our results on Computed Tomography (CT) images, with that of traditional graph-cut segmentation algorithm that is semi-manual (for a given organ).