|ECR 2015 / C-0553|
|High resolution imaging of scaphoid fracture with Herbert screw treatment: comparative findings on a cadaveric wrist with digital tomosynthesis, CT, CT with iterative reconstruction and cone beam CT|
|This poster is published under an open license. Please read the disclaimer for further details.|
Imaging techniques of the wrist
Trauma to the wrist affects all population groups and constitutes a significant public health issue because of the likelihood of fractures and ligament ruptures . Radiography remains the standard imaging tool but lacks sensitivity and specificity, and can lead to unnecessary wrist immobilization, particularly in patients with suspected scaphoid fracture. CT is more efficient but sometimes less available and can miss trabecular fractures . New imaging techniques have emerged that might be useful in the diagnosis of wrist injuries: digital tomosynthesis and cone beam CT (CBCT). The first is a pseudo-3D technique that is, today, mainly used in breast imaging . Cone beam CT (CBCT) is a fully 3D technique which recently became one of the standards in dentomaxillofacial imaging . Recent studies reported the additional value and opportunities of both technologies for the diagnosis of wrist pathology [1,4-6].
Imaging principles: 2D, pseudo-3D and 3D
Compared to 2D radiography, tomosynthesis is technique that improves the ability to visualize structures without the confusion of overlapping tissue. It is not a fully 3D technique as CT, but instead uses several radiography projections to reconstruct a pseudo-3D image in one direction. During the acquisition, multiple (20-40), low-dose projection images of the wrist are acquired at different angles, which are then reconstructed to produce a series of thin slices (Fig. 1a). Reviewing the wrist slice-by-slice with tomosynthesis reduces the confusion of superimposed tissue complexities as on a standard digital radiograph.
Compared to tomosynthesis, multislice computed tomography (CT) is a fully 3D technique which allows to reconstruct the area of interest in multiple planes. There is no confusion from superimposed tissue as with radiography and tomosynthesis. A high number of projections are reconstructed to a 3D image volume. CT uses a fan beam acquisition with multiple rotations at high speed (Fig. 1b). Different from CT, a CBCT acquisition involves a lower number of projections with a large cone shaped beam in a single, slow rotation, and uses a flat panel detector as in digital radiography (Fig. 1c). As with CT, the projection data are reconstructed to a 3D image volume. Due to their different technical characteristics, CT and CBCT images fundamentally result in images of different quality. The following four are key:
- High contrast resolution
The main benefit of CBCT is that it results in high resolution images with nearly isotropic voxel sizes (x,y,z) of 150 µm. This is achieved by a combination of specific hardware characteristis: limited scan field of view, small focal spot size and high resolution detector. High resolution CT typically results in 350 µm voxels that are not isotropic.
Due to its large cone beam and imaging volume during one rotation, CBCT is much more susceptible to scatter radiation. Scattered photons that are not detected by the small CT detector are now detected by the large flat panel detector with CBCT. Detected scatter-to-primary radiation ratio’s dramatically increase from about 0.2 with CT up to 0.4-3.0 with CBCT, depending on the imaging volume and beam energy. These higher scatter fractions will impair the image contrast capabilities of CBCT.
- Beam hardening
The phenomenon of beam hardening is the continuous shift of photon energy when the x-ray beam penetrates the body. This results in a shift of CT values (HU) for equal tissue type and possibly also in streak artifacts. Beam hardening is more pronounced in CBCT due to lower kilovolt and the higher heterogeneity of the x-ray beam.
- Acquisition speed
The slow tube rotation with CBCT (10-30s) compared to CT (<0.3s) limits its capabilities for dynamic studies and makes it more susceptible for moving artifacts.
The quality of 3D imaging of the scaphoid with osteosynthesis material is typically impaired due to a combination of lower spatial resolution and by the presence of metal which violates the reconstruction process (Fig. 2). Recent iterative reconstruction methods and the introduction of acquisition techniques with cone beam CT might help to improve 3D quality.
Thematically related posters
ECR 2015 / C-2635
Bone mineral density evaluation in patients with fractures according to digital radiography: assessment of risk of osteoporosis
ECR 2015 / C-2194
Comparison of sonographic and arthroscopic findings of trochlear cartilage of the knee joint
ECR 2015 / C-0704
Comparison of image optical density and accuracy of long bone shaft fractures obtained by digitization photographed and scanned radiographs.