Show simple item record

hal.structure.identifierCEntre de REcherches en MAthématiques de la DEcision [CEREMADE]
dc.contributor.authorFiot, Jean-Baptiste*
hal.structure.identifierCEntre de REcherches en MAthématiques de la DEcision [CEREMADE]
dc.contributor.authorCohen, Laurent D.
HAL ID: 738939
hal.structure.identifierCSIRO Information and Commuciation Technologies [CSIRO ICT Centre]
dc.contributor.authorRaniga, Parnesh*
hal.structure.identifierCSIRO Information and Commuciation Technologies [CSIRO ICT Centre]
dc.contributor.authorFripp, Jürgen*
dc.subjectBrain Imagingen
dc.subjectSupport Vector Machinesen
dc.titleEfficient Lesion Segmentation using Support Vector Machinesen
dc.typeCommunication / Conférence
dc.contributor.editoruniversityotherCSIRO Information and Commuciation Technologies (CSIRO ICT Centre) CSIRO;Australie
dc.description.abstractenSupport Vector Machines (SVM) are a machine learning technique that has been used for segmentation and classification of medical images, including segmentation of white matter hyper-intensities (WMH). Current approaches using SVM for WMH segmentation extract features from the brain and classify these followed by complex post-processing steps to remove false positives. The method presented in this paper combines the use of domain knowledge, advanced pre-processing (based on tissue segmentation and atlas propagation) and SVM classification to obtain efficient and accurate WMH segmentation. Features generated from up to four MR modalities (T1-w, T2-w, PD and FLAIR), differing neighbourhood sizes and the use of multi-scale features were compared. We found that although using all 4 modalities gave the best overall classification (average Dice scores of 0.54 ± 0.12, 0.72 ± 0.06 and 0.82 ± 0.06 respectively for small, moderate and severe lesion loads, using 3x3x3 neighbourhood intensity features); this was not significantly different (p = 0.50) from using just T1-w and FLAIR sequences (Dice scores of 0.52 ± 0.13, 0.71 ± 0.08 and 0.81 ± 0.07 for the same lesion loads and feature type). Furthermore, there was a negligible difference between using 5x5x5 and 3x3x3 features (p = 0.93). Finally, we show that careful consideration of features and preprocessing techniques leads to more efficient classification which outperforms the one based on all features with post-processing, and also saves storage space and computation time.en
dc.relation.ispartoftitleComputational Vision and Medical Image Processing: VipIMAGE 2011en
dc.relation.ispartofeditorJorge, R.M. Natal
dc.relation.ispartofeditorTavares, João Manuel R.S.
dc.relation.ispartofpublnameCRC Pressen
dc.subject.ddclabelTraitement du signalen
dc.relation.conftitleVIPIMAGE 2011, Third ECCOMAS Thematic Conference on Computational Vision and Medical Image Processingen

Files in this item


There are no files associated with this item.

This item appears in the following Collection(s)

Show simple item record