CT Prostate Segmentation Based on Continuously Updated Random Forests

DOI

10.2991/icmmita-15.2015.75How to use a DOI?

Keywords

Prostate Segmentation; Treatment Images; Continuously Updated Random Forests

Abstract

It is important to segment prostate automatically in the daily treatment images. However, previous methods often ignore the previous segmented treatment images which contain valuable patient-specific information. To this end, this paper proposes a novel CT prostate segmentation method based on a random forest model which is trained as a classifier to segment prostates. This model can be continuously updated by adding newly segmented prostate shapes into the training pool. In this way, more patient-specific information is incorporated into the training procedure. The experimental results show that the proposed method can improve the accuracy of prostate segmentation efficiently.

Copyright

© 2015, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Huangjian Deng
AU  - Xiubin Dai
AU  - Dandan Shi
PY  - 2015/11
DA  - 2015/11
TI  - CT Prostate Segmentation Based on Continuously Updated Random Forests
BT  - Proceedings of the 2015 3rd International Conference on Machinery, Materials and Information Technology Applications
PB  - Atlantis Press
SP  - 384
EP  - 389
SN  - 2352-538X
UR  - https://doi.org/10.2991/icmmita-15.2015.75
DO  - 10.2991/icmmita-15.2015.75
ID  - Deng2015/11
ER  -