TY - JOUR

T1 - CT and 3D-ultrasound registration for spatial comparison of post-EVAR abdominal aortic aneurysm measurements

T2 - A cross-sectional study

AU - Rouet, L

AU - Dufour, Cécile

AU - Collet Billon, Antoine

AU - Bredahl, K

PY - 2019

Y1 - 2019

N2 - OBJECTIVE: The aim of the present study is to provide a methodology to register volumes of stented abdominal aortic aneurysm, imaged by 3D-US and CT modalities. After registration, the method enables to compare the spatial location of measurements and AAA size in a common coordinate system.METHODS: The study is cross-sectional and compares volumes acquired within a few days, in order to eliminate changes due to the evolution of AAA shape after treatment. The key element is to rely on stent alignment to register the CT and 3D-US volumes, providing access to a patient-specific common spatial coordinate system. In parallel, 3D segmentations are performed and used to extract multi-planar reconstructions at the locations of maximum diameter in each modality. The positions of the planes extracted in each modality, and the AAA diameters are finally compared in the common coordinate system.RESULTS: Results are validated on a database of 52 patients. After registrations, results show a mean inter-planar distance of 6.4 ± 4.5 mm and a mean inter-planar angle of 10.2°±6.7 between CT and 3D-US multi-planar reconstructions. Bland-Altman comparisons of diameter measurements in the CT, US and non-registered volumes are respectively 5.1 ± 2.8, 3.9 ± 2.8, 4.6 ± 3.0 mm.CONCLUSION: The proposed approach provides both visual and quantitative validations of measurements extracted from multi-modality images of the same pathology, in terms of spatial relationship and diameters.SIGNIFICANCE: The present work provides additional confidence in the use of 3D-US without CT for the follow-up of patients with abdominal aortic aneurysms after endovascular treatment.

AB - OBJECTIVE: The aim of the present study is to provide a methodology to register volumes of stented abdominal aortic aneurysm, imaged by 3D-US and CT modalities. After registration, the method enables to compare the spatial location of measurements and AAA size in a common coordinate system.METHODS: The study is cross-sectional and compares volumes acquired within a few days, in order to eliminate changes due to the evolution of AAA shape after treatment. The key element is to rely on stent alignment to register the CT and 3D-US volumes, providing access to a patient-specific common spatial coordinate system. In parallel, 3D segmentations are performed and used to extract multi-planar reconstructions at the locations of maximum diameter in each modality. The positions of the planes extracted in each modality, and the AAA diameters are finally compared in the common coordinate system.RESULTS: Results are validated on a database of 52 patients. After registrations, results show a mean inter-planar distance of 6.4 ± 4.5 mm and a mean inter-planar angle of 10.2°±6.7 between CT and 3D-US multi-planar reconstructions. Bland-Altman comparisons of diameter measurements in the CT, US and non-registered volumes are respectively 5.1 ± 2.8, 3.9 ± 2.8, 4.6 ± 3.0 mm.CONCLUSION: The proposed approach provides both visual and quantitative validations of measurements extracted from multi-modality images of the same pathology, in terms of spatial relationship and diameters.SIGNIFICANCE: The present work provides additional confidence in the use of 3D-US without CT for the follow-up of patients with abdominal aortic aneurysms after endovascular treatment.

U2 - 10.1016/j.compmedimag.2019.02.004

DO - 10.1016/j.compmedimag.2019.02.004

M3 - Journal article

C2 - 30889540

VL - 73

SP - 49

EP - 59

JO - CMIGExtra: Cases

JF - CMIGExtra: Cases

SN - 1572-3496

ER -