Investigation of the role of feature selection and weighted voting in random forests for 3-D volumetric segmentation.
This paper describes a novel 3-D segmentation technique posed within the Random Forests (RF) classification framework. Two improvements over the traditional RF framework are considered. Motivated by the high redundancy of feature selection in the traditional RF framework, the first contribution deve...
Main Authors: | , , , |
---|---|
Format: | Journal article |
Language: | English |
Published: |
2014
|
_version_ | 1797090481089282048 |
---|---|
author | Yaqub, M Javaid, M Cooper, C Noble, J |
author_facet | Yaqub, M Javaid, M Cooper, C Noble, J |
author_sort | Yaqub, M |
collection | OXFORD |
description | This paper describes a novel 3-D segmentation technique posed within the Random Forests (RF) classification framework. Two improvements over the traditional RF framework are considered. Motivated by the high redundancy of feature selection in the traditional RF framework, the first contribution develops methods to improve voxel classification by selecting relatively "strong" features and neglecting "weak" ones. The second contribution involves weighting each tree in the forest during the testing stage, to provide an unbiased and more accurate decision than provided by the traditional RF. To demonstrate the improvement achieved by these enhancements, experimental validation is performed on adult brain MRI and 3-D fetal femoral ultrasound datasets. In a comparison of the new method with a traditional Random Forest, the new method showed a notable improvement in segmentation accuracy. We also compared the new method with other state-of-the-art techniques to place it in context of the current 3-D medical image segmentation literature. |
first_indexed | 2024-03-07T03:19:14Z |
format | Journal article |
id | oxford-uuid:b6de838d-5b6d-4f22-a2ea-91a6d17b6e1b |
institution | University of Oxford |
language | English |
last_indexed | 2024-03-07T03:19:14Z |
publishDate | 2014 |
record_format | dspace |
spelling | oxford-uuid:b6de838d-5b6d-4f22-a2ea-91a6d17b6e1b2022-03-27T04:44:08ZInvestigation of the role of feature selection and weighted voting in random forests for 3-D volumetric segmentation.Journal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:b6de838d-5b6d-4f22-a2ea-91a6d17b6e1bEnglishSymplectic Elements at Oxford2014Yaqub, MJavaid, MCooper, CNoble, JThis paper describes a novel 3-D segmentation technique posed within the Random Forests (RF) classification framework. Two improvements over the traditional RF framework are considered. Motivated by the high redundancy of feature selection in the traditional RF framework, the first contribution develops methods to improve voxel classification by selecting relatively "strong" features and neglecting "weak" ones. The second contribution involves weighting each tree in the forest during the testing stage, to provide an unbiased and more accurate decision than provided by the traditional RF. To demonstrate the improvement achieved by these enhancements, experimental validation is performed on adult brain MRI and 3-D fetal femoral ultrasound datasets. In a comparison of the new method with a traditional Random Forest, the new method showed a notable improvement in segmentation accuracy. We also compared the new method with other state-of-the-art techniques to place it in context of the current 3-D medical image segmentation literature. |
spellingShingle | Yaqub, M Javaid, M Cooper, C Noble, J Investigation of the role of feature selection and weighted voting in random forests for 3-D volumetric segmentation. |
title | Investigation of the role of feature selection and weighted voting in random forests for 3-D volumetric segmentation. |
title_full | Investigation of the role of feature selection and weighted voting in random forests for 3-D volumetric segmentation. |
title_fullStr | Investigation of the role of feature selection and weighted voting in random forests for 3-D volumetric segmentation. |
title_full_unstemmed | Investigation of the role of feature selection and weighted voting in random forests for 3-D volumetric segmentation. |
title_short | Investigation of the role of feature selection and weighted voting in random forests for 3-D volumetric segmentation. |
title_sort | investigation of the role of feature selection and weighted voting in random forests for 3 d volumetric segmentation |
work_keys_str_mv | AT yaqubm investigationoftheroleoffeatureselectionandweightedvotinginrandomforestsfor3dvolumetricsegmentation AT javaidm investigationoftheroleoffeatureselectionandweightedvotinginrandomforestsfor3dvolumetricsegmentation AT cooperc investigationoftheroleoffeatureselectionandweightedvotinginrandomforestsfor3dvolumetricsegmentation AT noblej investigationoftheroleoffeatureselectionandweightedvotinginrandomforestsfor3dvolumetricsegmentation |