TY - JOUR
T1 - Automatic Detection and Segmentation of Crohn's Disease Tissues From Abdominal MRI
AU - Mahapatra, Dwarikanath
AU - Schueffler, Peter
AU - Tielbeek, Jeroen
AU - Makanyanga, Jesica
AU - Stoker, Jaap
AU - Taylor, Stuart
AU - Vos, Frans
AU - Buhmann, Joachim
PY - 2013
Y1 - 2013
N2 - We propose an information processing pipeline for segmenting parts of the bowel in abdominal magnetic resonance images that are affected with Crohn's disease. Given a magnetic resonance imaging test volume, it is first oversegmented into supervoxels and each supervoxel is analyzed to detect presence of Crohn's disease using random forest (RF) classifiers. The supervoxels identified as containing diseased tissues define the volume of interest (VOI). All voxels within the VOI are further investigated to segment the diseased region. Probability maps are generated for each voxel using a second set of RF classifiers which give the probabilities of each voxel being diseased, normal or background. The negative log-likelihood of these maps are used as penalty costs in a graph cut segmentation framework. Low level features like intensity statistics, texture anisotropy and curvature asymmetry, and high level context features are used at different stages. Smoothness constraints are imposed based on semantic information (importance of each feature to the classification task) derived from the second set of learned RF classifiers. Experimental results show that our method achieves high segmentation accuracy with Dice metric values of 0.90 +/- 0.04 and Hausdorff distance of 7.3 +/- 0.8 mm. Semantic information and context features are an integral part of our method and are robust to different levels of added noise
AB - We propose an information processing pipeline for segmenting parts of the bowel in abdominal magnetic resonance images that are affected with Crohn's disease. Given a magnetic resonance imaging test volume, it is first oversegmented into supervoxels and each supervoxel is analyzed to detect presence of Crohn's disease using random forest (RF) classifiers. The supervoxels identified as containing diseased tissues define the volume of interest (VOI). All voxels within the VOI are further investigated to segment the diseased region. Probability maps are generated for each voxel using a second set of RF classifiers which give the probabilities of each voxel being diseased, normal or background. The negative log-likelihood of these maps are used as penalty costs in a graph cut segmentation framework. Low level features like intensity statistics, texture anisotropy and curvature asymmetry, and high level context features are used at different stages. Smoothness constraints are imposed based on semantic information (importance of each feature to the classification task) derived from the second set of learned RF classifiers. Experimental results show that our method achieves high segmentation accuracy with Dice metric values of 0.90 +/- 0.04 and Hausdorff distance of 7.3 +/- 0.8 mm. Semantic information and context features are an integral part of our method and are robust to different levels of added noise
U2 - https://doi.org/10.1109/TMI.2013.2282124
DO - https://doi.org/10.1109/TMI.2013.2282124
M3 - Article
C2 - 24058021
SN - 0278-0062
VL - 32
SP - 2332
EP - 2347
JO - IEEE Transactions on Medical Imaging
JF - IEEE Transactions on Medical Imaging
IS - 12
ER -