Findings and procedure details
UHRCT technologies
CT systems have developed from conventional CT,
helical CT,
MDCT,
into area detector CT (ADCT) to achieve wider,
faster,
and finer CT scanning with lower dose.
However,
CT in-plane spatial resolution has remained almost unchanged so far for approximately 30 years (Figure 3).
Eventually,
UHRCT has been newly introduced to remarkably improve in- and through-plane spatial resolution in March,
2017 (Figures 1 and 4).
Major features of UHRCT scanner include improved detector system (the minimal slice thickness,
0.25 mm; the maximal channel number,
1792) and x-ray focus (the smallest size,
0.4 x 0.5 mm) compared to a standard MDCT scanner (e.g.,
the minimal slice thickness,
0.5 mm; the maximal channel number,
896; the smallest x-ray focus size,
0.8 x 0.9 mm) (Figure 1).
The UHRCT detector represents the next breakthrough in detector design and manufacturing.
With 160 detector rows containing 1796 channels of only 0.25-mm thickness,
a new era in UHRCT imaging has begun (Figures 4 and 5).
To maximize the geometric efficiency of the detector,
the inter-septal gaps between the detector elements (Separator) were made much thinner,
maximizing the light sensitive area on the detector (Figure 5).
Using UHRCT scanner,
the following three scan modes can be performed: normal resolution (NR) mode,
high resolution (HR) mode,
and super high resolution (SHR) mode (Figure 6).
Voxel volume reconstructed in the SHR mode by UHRCT (matrix size,
1024 x 1024; slice thickness,
0.25 mm) can be reduced to one-eighth of that in the NR mode by UHRCT or that by standard MDCT (matrix size,
512 x 512; slice thickness,
0.5 mm) (Figure 6).
Reconstruction matrix size can be further increased from 1024 x 1024 to 2048 x 2048 in the HR and SHR modes,
and voxel volume can be simultaneously reduced to one-fourth (Figure 7).
Thus,
voxel volume reconstructed in the SHR mode by UHRCT can be ultimately reduced to 1/32 of that in the NR mode by UHRCT or that by standard MDCT.
Reconstruction algorithms
Since the invention of CT in the early 1970s,
CT images have been reconstructed from raw data using FBP algorithm.
Multiplying a negative value at both edges using a filter prior to back projections,
FBP reduces blurring compared to simple back-projection (SBP) algorithm (Figure 8).
Although image quality of FBP at lower radiation is rough and less optimal,
it is widely used owing to fast image reconstruction time and simplicity of method.
FBP and hybrid IR assume an infinitely small x-ray source or a focal spot that can be approximated by a point and assume that all x-ray photon interactions take place at a point located at the geometric center of the detector cell and not across the full area of the detector cell,
whereas hybrid IR is useful for reducing image noise by modeling system statistics compared to FBP.
On the other hand,
full IR (e.g.,
FIRST from Toshiba and Veo from GE) uses a forward reconstruction model and more precise integration of scanner geometry and the underlying physics (system optics) (Figure 9).
Here,
hybrid IR (e.g.,
AIDR3D from Toshiba,
ASIR from GE,
SAFIRE from SIEMENS,
iDose4 from Philips,
and Intelli IP from Hitachi) is the first generation iterative method applied by combining FBP and IR algorithm.
Increased spatial resolution by UHRCT increases image noise.
If required,
image noise can be reduced by applying a full IR algorithm (Forward projected model-based Iterative Reconstruction SoluTion [FIRST]; Toshiba Medical Systems).
This combination of UHRCT and FIRST has been newly introduced in October,
2017.
Various clinical applications
The combination of UHRCT and IR can improve delineation of fine structures and lesions of not only high contrast but lower contrast and diagnostic accuracy in various regions as follows: temporal bone (e.g.,
ossicles),
virtual bronchoscopy,
pulmonary ground-glass nodules,
subtle bony fractures,
and brain,
coronary,
and abdominal CT angiography for perforating brain arteries,
stented/calcified coronary vessel lesions,
and preoperative assessment for interventional radiology.
These clinical applications are described below.
Chest
UHRCT with a larger matrix size (1024 x 1024) improves the delineation of pulmonary nodules surrounded by ground-glass opacity on chest CT because of improved spatial resolution,
compared to standard MDCT (512 x 512) previously performed (Figures 10 and 11).
These nodules with the definitive diagnosis of lung adenocarcinoma more likely appear malignant by UHRCT.
The delineation of such nodules by UHRCT can be improved with the combined use of FIRST by remarkably reducing image noise (Figure 12).
Combined use of UHRCT and FIRST can also improve the delineation of peripheral bronchial trees (Figure 13).
In Figure 14,
CT virtual bronchoscopy by UHRCT with 0.25-mm slice thickness & 1024 matrix size was clinically useful for the diagnosis of lung cancer in segment 9 of the left lung by brush cytology under bronchoscopy compared to standard MDCT with 0.5-mm slice thickness & 512 matrix size.
In this patient,
the direct access of bronchoscopy to the target pulmonary nodule shown in yellow was difficult; however,
for the cytology examination,
the selection of the adequate bronchus extending to the nodule could be easily identified by UHRCT with delineating more distal bronchi than by standard MDCT.
Head and Neck
In temporal bone CT,
UHRCT is clinically feasible for improving spatial resolution and delineation of fine high-contrast structures such as the ossicles,
scutum,
tympanic tegmen,
and facial nerve canal as well as prostheses used for ossicular reconstruction,
since increased image noise by UHRCT is often much less problematic (Figure 15).
This image quality improvement by UHRCT in temporal bone CT (e.g.,
the stapes) is emphasized particularly on three dimensional (e.g.,
volume rendered [VR]) CT images (Figure 16).
In patients with cholesteatoma,
differentiation between soft tissue density and important bony structures such as the ossicles can be improved by UHRCT with improving spatial resolution,
compared to standard MDCT (Figure 17).
In a patient with tympanosclerosis associated with recurrent attic cholesteatoma,
use of FIRST combined with UHRCT better delineates faint and fine calcifications within soft tissue and is clinically useful for the accurate diagnosis owing to easy discrimination of the calcifications from image noise compared to FBP (Figure 18).
UHRCT is also clinically useful for paranasal and dental CT imaging.
In a patient with odontogenic maxillary sinusitis,
a radicular cyst,
the infection source,
is better delineated by UHRCT compared to standard MDCT (Figure 19).
Musculoskeletal
In Musculoskeletal CT,
UHRCT is also clinically feasible for improving delineation of subtle bony fracture,
since increased image noise by UHRCT is often much less problematic.
Actually,
in a patient with avulsion fracture of the proximal tibia (Figure 20) and with fracture of the major trapezoid bone (Figure 21),
the fracture and healing process are better delineated by UHRCT with 0.25-mm slice thickness & 1024 matrix size compared to standard MDCT.
Brain CT angiography
UHRCT with a larger matrix size (1024 x 1024 or 2048 x 2048) improves delineation of small peripheral branches in brain CT angiography (CTA) compared to standard MDCT (Figure 22).
This application is also clinically useful for better delineation of superficial cerebral veins on VR images with providing landmarks during an open surgery (Figure 23).
Combined use of FIRST with UHRCT can reduce image noise (Figure 24) and improving delineation and continuity of very small perforating arteries of approximately 0.1 to 0.5 mm in lumen diameter,
such as the anterior choroidal artery (Figures 26),
pontine artery (Figure 25),
and Heubner artery (Figure 27).
Cardiovascular
In coronary CTA,
improved spatial resolution and reduced partial volume effect by UHRCT are clinically beneficial for reducing blooming artifact from coronary artery calcifications (Figure 28) and stents (Figure 29) and improving delineation of the vessel lumen and non-calcified plaques,
if necessary,
combined with FIRST for reducing image noise (Figure 29).
Also in CT angiography of upper (Figure 30) and lower extremities (Figure 31),
small peripheral vessels and vascular lesions (e.g.,
hemangioma [Figure 30]) are better delineated by UHRCT (matrix size,
1024 x 1024; slice thickness,
0.25 mm) compared to standard MDCT (matrix size,
512 x 512; slice thickness,
0.5 mm).
Abdomen
Improved spatial resolution by UHRCT allows accurate measurement of CT attenuation value by reducing partial volume effect.
In a patient with adrenal adenoma (Figure 32),
use of UHRCT can more accurately detect subtle lipid within the tumor compared to standard MDCT by placing a region of interest within the tumor to measure the CT value (-3.4 HU vs.
11.8 HU).
In the same patient,
MR chemical shift images demonstrate the presence of lipid within the adrenal adenoma.
Note that the signal decreased from in-phase to opposed-phase imaging.
Use of UHRCT can provide CT virtual endoscopy of the gastrointestinal tract of excellent image quality like actual endoscopy.
Actually,
in a patient with early gastric cancer (Figure 33),
CT virtual endoscopy by UHRCT allows detailed assessment of the gastric mucosa and tumor.
In abdominal CT,
improved spatial resolution by UHRCT may be useful in some patients with pathologies of interest with relatively high contrast-to-noise ratio.
Actually,
in a patient with pancreatic neuroendocrine tumor (Figure 34),
use of UHRCT improves conspicuity of the well enhanced tumor on abdominal CT during the portal venous phase reconstructed with 5-mm slice thickness compared to standard MDCT previously performed.
Use of FIRST is remarkably useful for reducing image noise in lower contrast regions,
such as the abdomen and brain.
In the same patient,
on the abdominal CT reconstructed with 1-mm slice thickness,
use of FIRST markedly reduces image noise compared to FBP and improves tumor conspicuity compared to a hybrid IR (i.e.,
AIDR 3D) (Figure 34).
Use of UHRCT and FIRST can provide better preoperative assessment prior to interventional radiology as well as surgical operation,
particularly on abdominal CTA.
In a patient with hepatocellular carcinoma (HCC) (Figure 35),
use of UHRCT with larger matrix size better delineates feeding arteries to the HCC,
which is useful for adequate preoperative planning prior to transarterial chemoembolization (TACE).
Combined use of UHRCT and IR is clinically useful for markedly improving delineation of small peripheral branches of much better image quality on maximal-intensity-projection (MIP) and VR abdominal CTA (Figure 36).