Original title:
Tissue Characterisation In Spectral Ct Data
Authors:
Poláková, Veronika Document type: Papers
Language:
cze Publisher:
Vysoké učení technické v Brně, Fakulta elektrotechniky a komunikačních technologií Abstract:
This article deals with tissue characterisation in virtual monoenergetic images (VMI). It presents that with growing energy of VMI the median of CT number increases or decreases with different steepness depending on a type of tissue. As a consequence, some VMI enable better soft tissue distinction and therefore their better classification. To determine which VMI are best suited, Cohen d was used. After that, Random Forest classification algorithm was applied to these images. If median of pixels is considered in addition to pixels themselves, the tissues can be clasiffied correctly.
Keywords:
descriptive statistics; image segmentation; spectral CT; supervised machine learning Host item entry: Proceedings of the 25st Conference STUDENT EEICT 2019, ISBN 978-80-214-5735-5
Institution: Brno University of Technology
(web)
Document availability information: Fulltext is available in the Brno University of Technology Digital Library. Original record: http://hdl.handle.net/11012/186614