Reconstructing the human ribcage in 3d with x-rays and geometric models

Koehler,; Wischgoll,; Golshani,

doi:10.1109/mmul.2010.5692182

Cited by 4 publications

(2 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Several previous studies have proposed rib cage 3D reconstruction methods from PA (Postero-Anterior) and LAT (Lateral) X-ray views, which is the standard for clinical imaging and for spine 3D reconstruction (non-exhaustive list: Mitton et al 2000;Benameur et al 2001;Humbert et al 2009). Studies were dedicated either to computer aided diagnosis for chest analysis and lung disease (Dworzak et al 2010;Koehler et al 2010) or to subjectspecific FEM and scoliosis (Benameur et al 2005;Mitton et al 2008;Jolivet et al 2010;Seoud et al 2011;Grenier et al 2013). Statistical shape models, used in Benameur et al 2005 and Dworzak et al 2010, facilitate the fully automated detection of objects in images by providing a plausible shape with the respect of the shape variability extracted from a set of training objects.…”

Section: Introductionmentioning

confidence: 99%

3D reconstruction of rib cage geometry from biplanar radiographs using a statistical parametric model approach

Aubert

Vergari

Ilharreborde

et al. 2014

Computer Methods in Biomechanics and Biomedical Engineering: Im

View full text Add to dashboard Cite

is an open access repository that collects the work of Arts et Métiers ParisTech researchers and makes it freely available over the web where possible. This is an author-deposited version published in: https://sam.ensam.eu Handle ID AbstractRib cage 3D reconstruction is an important prerequisite for thoracic spine modelling, particularly for studies of the deformed thorax in adolescent idiopathic scoliosis. This study proposes a new method for rib cage 3D reconstruction from biplanar radiographs, using a statistical parametric model approach.Simplified parametric models were defined at the hierarchical levels of rib cage surface, rib midline and rib surface, and applied on a database of 86 trunks. The resulting parameter database served to statistical models learning which were used to quickly provide a first estimate of the reconstruction from identifications on both radiographs. This solution was then refined by manual adjustments in order to improve the matching between model and image. Accuracy was assessed by comparison with 29 rib cages from CT scans in terms of geometrical parameter differences and in terms of line-to-line error distance between the rib midlines. Intra and inter-observer reproducibility were determined regarding 20 scoliotic patients.The first estimate (mean reconstruction time of 2'30) was sufficient to extract the main rib cage global parameters with a 95% confidence interval lower than 7%, 8%, 2% and 4° for rib cage volume, antero-posterior and lateral maximal diameters and maximal rib hump, respectively. The mean error distance was 5.4 mm (max 35mm) down to 3.6 mm (max 24 mm) after the manual adjustment step (+3'30).The proposed method will improve developments of rib cage finite element modeling and evaluation of clinical outcomes.

show abstract

Section: Introductionmentioning

confidence: 99%

3D reconstruction of rib cage geometry from biplanar radiographs using a statistical parametric model approach

Aubert

Vergari

Ilharreborde

et al. 2014

Computer Methods in Biomechanics and Biomedical Engineering: Im

View full text Add to dashboard Cite

show abstract

“…The x-ray image is just a projection of the patient's body [7], where all the organs and bones are superimposed. So the surgeon has to mentally visualize the anatomy of interest [3].…”

Section: Introductionmentioning

confidence: 99%

3D Reconstruction of Bone Structures Based on Planar Radiography

Coelho

Ribeiro²,

Campos³

et al. 2015

Ambient Intelligence - Software and Applications

View full text Add to dashboard Cite

The 3D reconstruction of bone structures has many advantages in orthopedic applications. 3D bone models could be used in computer assisted surgery systems or in the pre-operative planning of an orthopedic surgery. The visualization of these models will lead to higher surgery accuracy. Usually the 3D reconstruction is done with CT or MRI scans. However these modalities have some disadvantages like the high costs, high acquisition time and high radiation. So, the planar radiography emerges as a more advantageous modality, because it avoids exposure to high radiation, reduces the acquisition time and costs and also is the most usually acquired study in the pre-operative planning of an orthopedic surgery. The principal challenge in reconstructing bone models from planar radiography is that a lot of information is missing when only one or two orthogonal images are used. So it's hard to obtain a precise geometry of the bone structure with only this information. In this work, we present a solution for the problem of reconstructing bone structures from planar radiography. With this solution, it's possible to obtain a 3D model of the bone that is suitable for orthopedic surgery planning.

show abstract

X2CT-GAN: Reconstructing CT From Biplanar X-Rays With Generative Adversarial Networks

Ying

Guo

Ma³

et al. 2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

156

119

View full text Add to dashboard Cite

Computed tomography (CT) can provide a 3D view of the patient's internal organs, facilitating disease diagnosis, but it incurs more radiation dose to a patient and a CT scanner is much more cost prohibitive than an X-ray machine too. Traditional CT reconstruction methods require hundreds of X-ray projections through a full rotational scan of the body, which cannot be performed on a typical X-ray machine. In this work, we propose to reconstruct CT from two orthogonal X-rays using the generative adversarial network (GAN) framework. A specially designed generator network is exploited to increase data dimension from 2D (X-rays) to 3D (CT), which is not addressed in previous research of GAN. A novel feature fusion method is proposed to combine information from two X-rays. The mean squared error (MSE) loss and adversarial loss are combined to train the generator, resulting in a high-quality CT volume both visually and quantitatively. Extensive experiments on a publicly available chest CT dataset demonstrate the effectiveness of the proposed method. It could be a nice enhancement of a low-cost X-ray machine to provide physicians a CT-like 3D volume in several niche applications.

show abstract

Reconstructing the human ribcage in 3d with x-rays and geometric models

Cited by 4 publications

References 13 publications

3D reconstruction of rib cage geometry from biplanar radiographs using a statistical parametric model approach

3D reconstruction of rib cage geometry from biplanar radiographs using a statistical parametric model approach

3D Reconstruction of Bone Structures Based on Planar Radiography

X2CT-GAN: Reconstructing CT From Biplanar X-Rays With Generative Adversarial Networks

Contact Info

Product

Resources

About