Many medical image analysis techniques require an initial localization and segmentation of anatomical structures. As part of the VISCERAL benchmarks on Anatomy segmentation, a hierarchical multi-atlas multi-structure segmentation approach guided by anatomical correlations is proposed. The method begins with a global alignment of the volumes and refines the alignment of the structures locally. The alignment of the bigger structures is used as reference for the smaller and harder to segment structures. The method is evaluated in the ISBI VISCERAL testset on ten anatomical structures in both contrast-enhanced and non-enhanced computed tomography scans. The proposed method obtained the highest DICE overlap score for some structures like kidneys and gallbladder. Similar segmentation accuracies compared to the highest results of the other methods proposed in the challenge are obtained for most of the other structures segmented with the method.
Anatomical correlations for a hierarchical multi-atlas segmentation of CT images
1. Anatomical correlations for a hierarchical
multi-atlas segmentation of CT images
Oscar A. Jiménez del Toro
University of Applied Sciences
Western Switzerland
(HES-SO)
4. Motivation
• Anatomical segmentation is fundamental for
further image analysis and Computer-Aided
Diagnosis1
• Manual annotation and visual inspection is
time consuming for radiologists
• Accurate large scale data analysis
techniques are needed
1 K.Doi. Current status and future potential of computer-aided diagnosis in medical
imaging. British Journal of Radiology, 78:3-19, 2005.
4
5. VISCERAL Benchmarks
• Automatic segmentation of
anatomical structures (20)
– Visceral Benchmark 1: 12
ceCT test volumes*10
structures
– ISBI challenge: 5 ceCT, 5
wbCT test volumes*10
structures
• CT and MR images
(contrast-enhanced and
non-enhanced)
8. Hierarchical multi-atlas segmentation
• Use multiple atlases for the
estimation on a target
image
• Global and local alignment
• Hierarchical selection of the
registrations improves
results2
• Label fusion
2Jiménez del Toro et.al., Multi-structure Atlas-Based Segmentation using Anatomical
Regions of Interest. In proceeding of: Medical Image Computing and Computer Assisted
Intervention (MICCAI2013) MCV workshop, Nagoya, Japan, 2013
14. Affine alignment
• Global
• Local refinement for
independent structures
• Regions of interest based on
the morphologically dilated
initial estimations
15. Affine alignment
• Global
• Local refinement for
independent structures
• Regions of interest based on
the morphologically dilated
initial estimations
16. Affine alignment
• Global
• Local refinement for
independent structures
• Regions of interest based on
the morphologically dilated
initial estimations
25. Conclusion
• Straightforward and fully automatic method
• Showed robustness in the segmentation of
multiple structures with high overlap for the
bigger structures (e.g. kidneys, liver, lungs)
• Smaller structures fared well compared to the
other approaches
• Future work:
– Extend to method to other modalities (CTwb ISBI challenge, MR)
– Improve speed of the algorithm