Feature extraction and classification for rectal bleeding in prostate cancer radiotherapy: A PCA based method

Abstract : In this work, we studied the efficiency of principal component analysis for feature extraction and classification of prostate cancer patients suffering from rectal bleeding. We fully exploited the three-dimensional planned dose distribution by considering the voxels as observations. We compared different possibilities for selecting the most relevant features (sequential and combinatory). The receiving operator characteristics were used as performance criterion. The obtained results demonstrate the ability of the method to classify two groups of patients, namely rectal bleeding and non-rectal bleeding. They also suggest that local dose/toxicity relationships exist.
Document type :
Journal articles
Complete list of metadatas

Cited literature [6 references]  Display  Hide  Download

https://hal-univ-rennes1.archives-ouvertes.fr/hal-00956765
Contributor : Aureline Fargeas <>
Submitted on : Friday, March 7, 2014 - 3:45:30 PM
Last modification on : Thursday, February 28, 2019 - 3:53:59 PM
Long-term archiving on : Saturday, June 7, 2014 - 11:25:27 AM

File

2013-IRBM-Farg-PCA.pdf
Files produced by the author(s)

Identifiers

Citation

Aureline Fargeas, Amar Kachenoura, Oscar Acosta, Laurent Albera, Gael Drean, et al.. Feature extraction and classification for rectal bleeding in prostate cancer radiotherapy: A PCA based method. IRBM, Elsevier Masson, 2013, 34 (4-5), pp.296-299. ⟨10.1016/j.irbm.2013.07.009⟩. ⟨hal-00956765⟩

Share

Metrics

Record views

1330

Files downloads

311