Abstract
The purpose of this paper is to introduce the Full-Body Image segmentation tool, FBIseg, for Visible Human Project and similar projects around the world. Currently, full body image segmentation is typically performed manually with different level of computer assistance. It is laborious and prohibitive. The FBIseg developed by our group combines the state-or-the-art machine segmentation capability with human expert knowledge to attain efficiency and accuracy in full-body segmentation. The FBIseg tool was written in the Java programming language for its portability. Key features in the FBIseg tool include the use of multiple segmentation techniques on a dataset (e.g. CT scan of a single subject) and the switching of segmentation algorithms from one slice image to another for optimizing segmentation efficiency and accuracy.
Original language | English |
---|---|
Pages (from-to) | S40 |
Number of pages | 1 |
Journal | International Journal of CARS |
Volume | 9 |
Issue number | Supp 1 |
DOIs | |
Publication status | Published - May 2014 |
Event | 28th International Congress and Exhibition on Computer Assisted Radiology and Surgery - Fukuoka, Japan Duration: 25 Jun 2014 → 28 Jun 2014 Conference number: 28th |