Results

Method details

Download PDF file of article

Abstract:
This paper describes an interactive approach to the identification of coronary arteries in 3D angiography images that allows for a complete segmentation with little manual labor. The approach is based on a novel multiple hypothesis tracking methodology which is complemented with a standard minimal path search. When evaluated using the 3D CT angiography data supplied with the MICCAI’08 workshop 3D Segmentation in the Clinic: A Grand Challenge II, all target coronary artery centerlines could be identified in about 5 minutes per data set. A 98 % overlap with expert human manual segmentations was achieved as well as a subvoxel spatial accuracy within the expert inter-observer variability.


Summary

Measure% / mmscorerank
 min.max.avg.min.max.avg.min.max.avg.
OV 94.5% 100.0% 99.3% 49.4 100.0 91.6 1 1 1.00
OF 23.9% 100.0% 94.6% 12.7 100.0 83.2 1 1 1.00
OT 94.5% 100.0% 99.5% 48.9 100.0 90.0 1 1 1.00
AI 0.16 mm 0.37 mm 0.24 mm 38.5 60.4 47.9 1 1 1.00
Total             1 1 1.00


All results

  OVOFOTAIAvg.
DV%scorerank%scorerank%scorerankmmscorerankrank
0 0 100.0 100.0 1 100.0 100.0 1 100.0 100.0 1 0.37 42.3 1 1.0
0 1 98.5 86.7 1 88.2 48.4 1 98.5 83.8 1 0.33 44.2 1 1.0
0 2 99.8 95.3 1 97.5 96.5 1 99.8 95.1 1 0.31 52.6 1 1.0
0 3 100.0 100.0 1 100.0 100.0 1 100.0 100.0 1 0.32 43.6 1 1.0
1 0 100.0 100.0 1 100.0 100.0 1 100.0 100.0 1 0.22 47.0 1 1.0
1 1 100.0 100.0 1 100.0 100.0 1 100.0 100.0 1 0.23 50.9 1 1.0
1 2 100.0 100.0 1 100.0 100.0 1 100.0 100.0 1 0.24 45.4 1 1.0
1 3 100.0 100.0 1 100.0 100.0 1 100.0 100.0 1 0.25 50.4 1 1.0
2 0 100.0 100.0 1 100.0 100.0 1 100.0 100.0 1 0.16 51.9 1 1.0
2 1 100.0 100.0 1 100.0 100.0 1 100.0 100.0 1 0.22 45.2 1 1.0
2 2 100.0 100.0 1 100.0 100.0 1 100.0 100.0 1 0.16 54.5 1 1.0
2 3 100.0 100.0 1 100.0 100.0 1 100.0 100.0 1 0.26 39.8 1 1.0
3 0 100.0 100.0 1 100.0 100.0 1 100.0 100.0 1 0.18 60.4 1 1.0
3 1 100.0 100.0 1 100.0 100.0 1 100.0 100.0 1 0.32 47.7 1 1.0
3 2 94.5 87.8 1 93.8 93.4 1 94.5 87.8 1 0.21 50.6 1 1.0
3 3 100.0 100.0 1 100.0 100.0 1 100.0 100.0 1 0.32 48.2 1 1.0
4 0 96.1 84.6 1 94.9 84.7 1 100.0 100.0 1 0.16 44.8 1 1.0
4 1 96.0 49.7 1 23.9 12.7 1 96.0 49.0 1 0.18 53.0 1 1.0
4 2 100.0 100.0 1 100.0 100.0 1 100.0 100.0 1 0.24 42.7 1 1.0
4 3 98.7 62.8 1 95.0 47.5 1 100.0 100.0 1 0.17 51.4 1 1.0
5 0 100.0 100.0 1 100.0 100.0 1 100.0 100.0 1 0.31 55.1 1 1.0
5 1 96.9 49.4 1 86.0 47.7 1 96.9 48.9 1 0.30 46.2 1 1.0
5 2 100.0 100.0 1 100.0 100.0 1 100.0 100.0 1 0.29 46.5 1 1.0
5 3 99.9 82.8 1 99.8 50.0 1 99.9 50.0 1 0.28 51.9 1 1.0
6 0 100.0 100.0 1 100.0 100.0 1 100.0 100.0 1 0.16 45.6 1 1.0
6 1 99.8 95.7 1 99.5 49.8 1 99.8 49.9 1 0.17 50.2 1 1.0
6 2 100.0 100.0 1 100.0 100.0 1 100.0 100.0 1 0.25 50.5 1 1.0
6 3 100.0 100.0 1 100.0 100.0 1 100.0 100.0 1 0.16 48.6 1 1.0
7 0 99.7 82.0 1 91.3 46.5 1 100.0 100.0 1 0.19 46.2 1 1.0
7 1 99.7 72.5 1 88.5 73.5 1 99.7 72.3 1 0.25 42.4 1 1.0
7 2 98.8 88.2 1 75.0 37.8 1 98.8 49.7 1 0.29 38.5 1 1.0
7 3 99.8 93.7 1 94.0 73.6 1 99.8 93.7 1 0.26 44.3 1 1.0