2019
DOI: 10.1007/978-3-030-32226-7_70
|View full text |Cite
|
Sign up to set email alerts
|

Towards Fully Automatic X-Ray to CT Registration

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
4
1

Citation Types

0
15
0

Year Published

2020
2020
2021
2021

Publication Types

Select...
4
2
1

Relationship

1
6

Authors

Journals

citations
Cited by 30 publications
(15 citation statements)
references
References 9 publications
0
15
0
Order By: Relevance
“…Pelvis poses were estimated using these annotations, yielding reprojection errors of 14-34 mm for other landmarks not learned by the network. Their work was extended in [20], whereby each network was fine-tuned on simulated fluoroscopy for a specific patient of interest. The approach was evaluated by estimating landmark locations in previously unseen simulated images, and using these estimates to produce quality initializations for 2D/3D registration.…”
Section: Introductionmentioning
confidence: 99%
See 1 more Smart Citation
“…Pelvis poses were estimated using these annotations, yielding reprojection errors of 14-34 mm for other landmarks not learned by the network. Their work was extended in [20], whereby each network was fine-tuned on simulated fluoroscopy for a specific patient of interest. The approach was evaluated by estimating landmark locations in previously unseen simulated images, and using these estimates to produce quality initializations for 2D/3D registration.…”
Section: Introductionmentioning
confidence: 99%
“…The approach was evaluated by estimating landmark locations in previously unseen simulated images, and using these estimates to produce quality initializations for 2D/3D registration. No analysis on actual fluoroscopy was conducted in [20].…”
Section: Introductionmentioning
confidence: 99%
“…The dataset size also varied between 20–40 cases, 50–70 cases, and more than 100 cases . Some studies used the pure 2D CNN, others used the 2D CNN for the 2D representations of 3D volumes, and still others used the 3D CNN directly . The image ROI in these studies were significantly different, including the head, pelvis, distal femur, and brain .…”
Section: Discussionmentioning
confidence: 99%
“…Machine learning methods have achieved a major success in the facial landmark detection in common 2D images, as well as the anatomical landmark detection in 2D and 3D medical images . However, automatic landmarking in cephalometric analysis has not yet fully benefited from this progress, and few studies have been conducted in this area .…”
Section: Introductionmentioning
confidence: 99%
“…It was shown in [5] that it is possible to generate an accurate, scaled 3D model of a patient's pelvis based on a single anterior-posterior radiograph; however, because only a single shot was used and no object of known size was present in the image, this technique was unable to resolve the scale. More recently, it was shown that it is possible to identify anatomical landmarks on intraoperative X-rays using a Convolutional Neural Network [3], which can assist the surgeons in better estimating the 3D position of the desired anatomy based on the 2D radiographs. In this study, we propose to train a network on a pre-operatively available Computed Tomographic (CT) volumeset and to use it to recognize the position of the C-Arm based on a single X-ray image.…”
Section: Introductionmentioning
confidence: 99%