The Right Scanner Parameters to

Andrea Laghi and Pasquale Paolantonio

CONTENTS

6.1 Introduction 61

6.2 Collimation 62

6.3 Image Reconstruction 64

6.4 Pitch 64

6.5 Tube Current Setting and Low Dose Protocols 65

6.6 Practical Guidelines 68 Appendix. Glossary of Terms 69 References 70

Introduction

The advent of CT colonography (CTC) (or, virtual colonoscopy) in 1994 was made possible by the development of spiral CT technology, which provides a volumetric coverage of a cleansed and air-distended colon within a single breath-hold (Vining et al. 1994).

The introduction of multidetector row computed tomographic (MDCT) scanners in late 1998 opened a new era for CT in general and CTC in particular (Berland and Smith 1998). The use of multiple detector arrays along the z-axis offers substantial benefits related to anatomic coverage, scanning time and longitudinal spatial resolution compared with single-slice spiral CT (SSCT) (Beaulieu et al. 1998; Fenlon et al. 1999; Hara et al. 2001).

In principle, using similar parameters on both SSCT and MDCT results in wider anatomic coverage and faster scanning time with MDCT. On the other hand, MDCT provides sub-millimeter collimation, improves z-axis resolution and generates isotropic voxels, thus resulting in better image quality of reformatted planes as well as three-dimensional reconstructions. The drawback is represented by data explosion, with generation of over 1000 images per scan per patient, making data viewing and analysis ("data workflow") a key issue to be solved within the next period of time (Sherbondy et al. 2005).

Another major issue is represented by dose delivery, usually higher with CTC compared with a standard abdominal CT study due to routine use of prone and supine scans (Chen et al. 1999). Radiation exposure has also substantially increased over the past few years due to the widespread use of thinner collimations and the consequent increase of tube current setting in order to reduce image noise. Low or ultra-low dose MDCT protocols together with new automatic dose modulation software may help in solving this problem representing a crucial issue for proposing VC as a screening method for colonic polyp in healthy subjects (Iannaccone et al. 2003a).

A further variable in CTC scanning parameters is due to technological differences among CT vendors and MDCT generations. It does exist quite a large experience on investigations using 4-slice MDCT scanners, but, on the other hand, there are only few manuscripts on 16-slice MDCT and almost no experience, at the time of this paper, on new 64-slice scanners. As technology continues to advance, there will be a continuing need to reassess the relative tradeoffs between scan width, image noise, patient dose, image artefacts, breath-hold times, and the number of reconstructed images to be viewed and archived.

As a consequence of what is discussed above, the "panorama" of technical approaches is expanding, offering a wide spectrum of different possibilities. For these reasons, the answer to the question "which are the right parameters to be used for CTC?" may be puzzling.

Department of Radiological Sciences, University of Rome "La Sapienza", Polo Didattico Pontino, I.C.O.T., Via Franco Fag-giana 34, 04100 Latina, Italy

Collimation

Collimation is the parameter that - more than the others - has dramatically changed since the development of CTC in parallel with continuous evolution of scanner technology.

"Thin" collimation is a mandatory pre-requisite for a CTC study. The question is how thin is a thin collimation? The answer may be either political or technical. In fact, collimation is strictly related to the size of the target lesion. Since on CT you are not able to detect a lesion smaller than the effective slice thickness due to partial volume effect, the size of the ideal target lesion should be defined. This should avoid the risk of searching for the thinnest possible collimation as soon as a new equipment becomes clinically available.

On a technical point of view, a consensus was achieved (Barish et al. 2005). If considering SSCT, the maximum accepted collimation is 5 mm with overlapped image reconstruction (usually set at 3 mm) (Hara et al. 1997a). In the first CTC studies, due to limited tube capacity, two or three consecutive breath-holds were needed to cover the entire colon. With the progress in tube technology a single breath-hold of 40-50 s was made possible on SSCT scanners (Taylor et al. 2003).

One of the major benefits of MDCT was represented by the possibility of further reducing the collimation compared with SSCT. Although some authors, at the beginning of the era of MDCT, still proposed 5 mm as an optimal collimation due to acquisition time, breathing artefacts as well as image workload (Hara et al. 2001), it is now widely accepted that a collimation no thicker than 3 mm is mandatory (Barish et al. 2005).

A major drawback of the use of thin collimation is the increased current tube setting, necessary in order to reduce image noise and to maintain an acceptable image quality. This is the reason why several researchers have performed investigations on the potential benefits of thin collimation protocols (Whiting et al. 2000; Fletcher et al. 2000; Rogalla and Meiri 2001; Gillams and Lees 2002; Macari et al. 2002; Taylor et al. 2003; Wessling et al. 2003) looking for the clinical impact of thin col-limation protocols in terms of polyp detection and characterization.

Several in-vitro evaluations were reported. Although phantom models have inherent limitations represented by the ideal conditions of the study design (a true colon is a "moving" organ, with peri staltic motion as well as motion related to the anatomic location, partly intra-peritoneal and partly retroperitoneal) they may provide useful data to be tested on patients.

In a personal experience (Laghi et al. 2003), we built a phantom model with 12 lesions ranging in size between 3.2 mm and 12 mm; lesion morphology simulated sessile polyps and flat and depressed lesions. Different scanning protocols were compared based on 1.0-mm, 2.5-mm and 5-mm collimations with effective slice thicknesses of respectively 1.25 mm, 3 mm and 5 mm. Results showed the best performance when using 1.0-mm collimation protocol (no lesions missed), although a statistically significant difference was observed only among 1.0/2.5-mm protocols in comparison with 5.0-mm protocol. If only lesions larger than 10 mm were considered no differences were observed among the three different protocols (Fig. 6.1).

In another study (Wessling et al. 2003), a phantom with simulated polypoid lesions ranging in size between 2 mm and 12 mm was built. Different protocols were compared with collimation ranging between 1.25 mm and 5 mm. Results showed no significant differences for detection of lesions larger than 10 mm, but for lesions smaller than 6 mm a clear benefit was observed when a thin collimation protocol (1.25 mm) had been used.

Finally, in another in-vitro experience (Taylor et al. 2003), where a human colonic specimen of a patient with familiar adenomatous polyp syndrome after total colectomy was used as a phantom, colli-mation had a significant effect on the polyp detection rate which, when unadjusted for size, was 50% higher at 1.25 mm collimation than at 2.5 mm. This effect was most marked for polyps less than 5 mm, for which a small but significant improvement in detection also occurred with increased tube current. For polyps 5 mm or larger the benefit of decreased collimation was significant resulting in 7% improved detection rate. The use of thin colli-mation is also associated with a possible increase of specificity, especially of small lesions, since the detection of tiny air bubbles might be easier.

Although thin collimation protocols may provide benefits in terms of detection of small polypoid lesions, the application to patients is limited by technical restrictions of four-slice MDCT. In fact, with four-slice MDCT a compromise between scanning time and collimation is necessary since the use of 1 mm is associated with scanning time over 30 s (Fig. 6.2). On the other hand, the increased radiation exposure as compared with thicker collimation a

Fig. 6.1a,b. Colonic phantom containing three different simulated lesions: 9.6-mm and 5.5-mm sessile polyps and 8-mm "flat" lesion. Using: a a thick collimation protocol lesions' sharpness is definitely reduced compared with; b a thin collimation protocol. Note edge blurring (arrow) directly related to the increase of effective slice thickness as well as geometric distortion. This artefact particularly affects simulated "flat" lesion (arrowhead)

25mm Polyp During Colonoscopy

Fig. 6.2a,b. Four-slice MDCT scan of the abdomen and pelvis: "high-resolution" protocol with 1-mm collimation and1.25 mm effective slice thickness vs "fast scanning" protocol with 2.5-mm collimation and 3 mm effective slice thickness (b). The quality of coronal reformatted image is much improved on high-resolution scanning protocol although the acquisition time is much longer

Fig. 6.2a,b. Four-slice MDCT scan of the abdomen and pelvis: "high-resolution" protocol with 1-mm collimation and1.25 mm effective slice thickness vs "fast scanning" protocol with 2.5-mm collimation and 3 mm effective slice thickness (b). The quality of coronal reformatted image is much improved on high-resolution scanning protocol although the acquisition time is much longer protocols as well as the unclear benefits of detecting small polypoid lesions (smaller than 5 mm) generated controversies among different research groups. As an example, some authors reported the benefits of 1-mm collimation (Macari et al. 2002) whereas others recommended 2.5 mm (effective slice thickness, 3 mm) (Power and Pryor 2002) or even 5 mm (Hara et al. 2001; McCollough 2002). The theoretical effect of changing collimation on polyp detection was also investigated by reconstructing 1-mm data sets at various section thicknesses; sensitivity for polyps measuring 3-5 mm decreased from 96% at a 1 mm section thickness to 74% at a 5 mm section thickness (Rogalla and Meiri 2001).

To summarize, data from both phantoms and patient populations suggest that, for four-slice MDCT, a collimation no larger than 3 mm should be implemented. Thinner collimations may provide benefits in terms of detection of small polypoid lesions (smaller than 5 mm) although the effects on false positive rates as well as on radiation exposure have not been fully evaluated.

First results on 16-slice MDCT scanners confirm these data, suggesting a scanning protocol with 1mm collimation for the detection of lesions smaller than 3 mm and with 3-mm collimation for lesions larger than 3 mm (Rottgen et al. 2003).

The advent of new 64-slice MDCT has completely solved the scanning problems, since no compromise between collimation and scan time is necessary any more (Fig. 6.3). Collimation is routinely ranging between 1.2 mm and 0.6 mm and scanning time is below 10 s (Flohr et al. 2005). Results are expected in the next few months, with a possible consistent improvement in the detection of tiny polyps. Problems to be face with are image noise of sub-millimetre collimation protocols, dose delivery and the huge datasets needing powerful workstations to be managed. At the time of this chapter, there is not enough experience to answer these raising issues.

Image Reconstruction

Resolution in the z-direction is particularly important for three-dimensional reconstructions. With SSCT a compromise between slice thickness, anatomical coverage and scanning time is necessary. The result is a relative thick collimation (3-5 mm) and consequently slice thickness negatively affecting the quality of reconstructed images. A possible solution, in order to improve z-axis resolution using SSCT, is to reconstruct raw data in order to obtain overlapped slices. Usually, 60% overlap is recommended (Paul et al. 1999).

Follicles Bearded Dragon
Fig. 6.3. Tissue Transition Projection (TTP) (or "virtual double contrast enema") showing the entire colon scanned with sub-millimetre collimation during an acquisition time of less than 10 s

The technical approach with MDCT is completely different. The reconstruction process for MDCT makes use of multiple-row data collection to minimize image degradation induced by rapid patient translation using reconstruction algorithms (Taguchi and Aradate 1998; Flohr et al. 2005). MDCT makes use of data interpolation and different reconstruction algorithms in order to obtain from an original data set of images acquired with a fixed collimation different data sets with different slice thickness (Hu 1999; Hu et al. 2000). Of course, the reconstructed images cannot be thinner than detector collimation. With this approach the reconstruction of overlapped slices is no longer recommended as for SSCT.

6.4 Pitch

Pitch is a technical parameter intrinsic to spiral CT correlating table travel, collimation and gantry rotation time. In other terms, pitch=TS RT/C, where TS is table speed, expressed in mm/s, RT is gantry rotation time, expressed in seconds and C is collima-tion expressed in mm. Pitch directly affects image quality, since the higher the pitch value the broader the Slice Sensitivity Profile (SSP) (see Appendix). It means that the effective scan width increases with increasing pitch. High pitch values introduce artefacts especially along the z-axis, resulting in blurred images. For this reason, a maximum pitch value of 2 is advisable.

In colon imaging, spiral artefacts induced by high pitch values may affect, in particular, those structures changing dramatically the anatomic position along the z axis (i.e., hepatic and splenic flexures, sigmoid colon) as well as polyp visualization. This is the case of rippling artefacts which are introduced in the images when increasing angle relative to the z axis as well as when pitch is increased; rippling artefacts result in a negative effect on the depiction of sessile polyps (Whiting et al. 2000) (Fig. 6.4).

Moreover, the degree to which partial volume effects distort polyp morphology is determined by polyp size relative to the effective section thickness and section sensitivity profile, which are primarily a function of collimation, pitch and interpolation algorithm. For SSCT protocols with 3 mm and 5 mm collimation and pitch ranging between one and two, excellent depiction of 6-mm and 13-mm peduncu-lated polyps and 10 3 mm sessile polyps can be

Fig. 6.4. Example of rippling artefacts induced by high pitch value. The small sessile polyp in a simulated colonic phantom (arrow) is barely seen

achieved. With a protocol including 5-mm collima-tion and pitch two, flat lesions of 11 mm 1 mm and 6 mm 2 mm can be degraded by rippling artefact (Whiting et al. 2000).

It has been observed that polyp conspicuity decreases when pitch increases, in particular on three-dimensional reconstructions (Dachman et al 1997).

In a different study using a pig colon phantom the prevalence of adverse CT artefacts over a range of scanning parameters was assessed, demonstrating that smoothing becomes more evident when collimation increases, whereas stair-step artefacts and longitudinal distortion are more dependent on increasing pitch (Springer et al. 2000).

Pitch is also correlated to dose delivery to patients, expressed in mGy, as CT Dose Index (CTDI) (see

Appendix).

CTDI is inversely proportional to pitch (e.g. doubling pitch halves the CTDI). This is true only if mA value remains constant. This is not the case for all manufacturers, since in some scanners mA is automatically adjusted for pitch so CTDI remains constant.

The introduction of MDCT has also modified the concept of pitch, which can be defined either relative to total X-ray collimation (pitchx) or to individual detector width (pitchd). Consequently pitchd=pitchx number of slices. As an example, on 4-slice MDCT, with 12.5 mm/s table speed of 2.5 mm collimation, pitchd is 2.5 (12.5 mm/s/2.5 mm 0.5 sec) whereas pitchx is 0.625 (12.5 mm/s/10 mm 0.5 s) where

10 mm is total X-ray beam (2.5 mm 4). For the sake of clarity and uniformity, the detector pitch should no longer be used. For both single-section CT and multi-detector row CT, the pitch (p) is given by p=TF/w, where TF is table feed per rotation, and w is total width of the collimated beam, according to International Electrotechnical Commission specifications. For p<1, data acquisition occurs with overlap in the z-axis direction; for p>1, data acquisition occurs with gaps.

Some remarkable differences about the influence of pitch on image quality between SSCT and MDCT do exist. In fact, on MDCT the influence of pitch on image quality depends on the scanner. Some manufacturers recommend particular pitch values, whilst others allow any pitch to be used. The difference depends on data interpolation properties. However, in general, MDCT allows the use of faster table speed compared with SSCT.

On a human colonic specimen phantom, for polyps of 5 mm or larger the effect of pitch on detection is insignificant on MDCT using pitch value ranging from 0.75 to 1.5 (Taylor et al. 2003). However this is not true for small lesions, smaller than 5 mm for whom the detection rate is higher with lower pitch value.

Finally, although pitch in MDCT is a less essential parameter in terms of section sensitivity profile, image noise in MDCT depends on pitch (for fixed tube current-time product in mAs). This is opposite to SSCT scanning where image noise is virtually pitch-independent. To account for that difference, MDCT systems automatically increase mA/s as the pitch is increased to maintain comparable image noise.

Tube Current Setting and Low Dose Protocols

One of the major limitations preventing CTC from being used in screening programs for patients at risk for colorectal carcinoma is its relatively high radiation exposure. There are three main reasons that account for the high radiation dose of CTC. First, the technique is usually performed in the prone and supine positions, because both positions have been found to detect the highest number of lesions; this, of course, doubles the radiation dose to the patient (Chen et al. 1999). Second, CTC examinations are currently performed with MDCT scanners, which tend to have a higher effective dose level for the same dose compared with SSCT, due to geometric inefficiency (Giacomuzzi et al. 2001). This is especially true if considering 4-row scanners, although geometric efficiency has much improved on 16- and 64-MDCTs. Third, there is a trend to use narrower collimations (1.0 mm or even less instead of 2.5 mm or 5.0 mm); this has the great advantage of near isotropic spatial resolution (i.e., the voxels have almost identical sides along the three axes) but, at the same time, leads to an increase in effective dose (Van Gelder et al. 2002).

In a single experience, a higher number of small polyps (<5 mm) was detected using thin collima-tion (1.25 mm), low pitch, and high tube current (150 mA) (Taylor et al. 2003). Unfortunately, the associated dose penalty was prohibitive, with an effective dose of 20.0 mSv for combined supine and prone scanning. Strategies to reduce patient radiation exposure although working at thin collimation include high pitch value as well as low tube current. Thus doubling the pitch and reducing the tube current setting to 50 mA would deliver an effective dose for combined supine and prone scans of 3.4 mSv, lower than the dose of either a standard abdomino-pelvic CT scan (6-24 mSv) or a barium enema study (6.4 mSv).

Indeed, in a recent study, Van Gelder et al. showed that the median effective dose for complete (i.e., prone and supine acquisitions combined) CTC in 12 different institutions is about 8.8 mSv. CTC at 8.8 mSv may result in a risk of up to 0.02% for inducing cancer in the population over 50 years (who are currently considered the target population for colorectal cancer screening) (Van Gelder et al. 2002). Considering these factors, increasing attention has been focused on the optimization of low-dose protocols for CTC.

In theory, in view of the inherently high contrast between the air-filled lumen of the colon and the soft-tissue attenuation of the colonic wall, a relevant dose reduction should be achievable without loss of diagnostic accuracy. The low-dose technique is associated with an increase in image noise because of the reduced number of X-rays reaching the detectors. The increase in image noise may deteriorate lesion conspicuity in terms of detectability on low-contrast imaging. The noise level on CT images depends on the accuracy of the transmission measurements used in the reconstruction of the images, which in turn depends primarily on the number of transmitted photons. This number is proportional to the dose used for the CT scan and also depends strongly on the size of the patient.

The imaging of high-contrast structures allows a higher noise level with a satisfactory image quality and diagnostic reliability. Thus, it seems feasible to decrease the dose in CTC because endolumi-nal lesions show an inherently high contrast to the surrounding colonic air or gas (Figs. 6.5 and 6.6). Because subjectively perceived noise is inversely related to window width, image analysis is usually performed with a wide window centre and setting.

The first attempt was performed in 1997, when it was proposed to reduce data size and radiation dose delivery for SSCT colonography (Hara et al. 1997b), using 70 mA.

Subsequently, the same authors took advantage of the faster data acquisition provided by MDCT and demonstrated better bowel distension and fewer respiratory artefacts with a beam collimation (5.0 mm) and an effective radiation dose (4.7 mSv for men; 6.7 mSv for women) comparable to those of SSCT (Hara et al. 2001). A different proposed approach was to use a thin (i.e., 1.0 mm) beam collimation protocol to obtain images with near isotropic voxels, but simultaneously decrease the effective mAs to 50 in order to keep the effective radiation dose to a level comparable to that of SSCT (5.0 mSv for men and 7.8 mSv for women) (Macari et al. 2002). Such a protocol provided excellent sensitivity for detection

Fig. 6.5. Low dose scan of normal sigmoid colon. Due to inherently high contrast between colonic surface and endoluminal air, image quality of colonic lumen is still optimal, in particular if image analysis is performed with a wide window centre and setting

Fig. 6.6a,b. The increase in image noise is perceived more on three-dimensional endoluminal views: a, standard dose scan; b, low dose scan. Note sharpness of endoluminal surface negatively affected by dose reduction in b

Fig. 6.6a,b. The increase in image noise is perceived more on three-dimensional endoluminal views: a, standard dose scan; b, low dose scan. Note sharpness of endoluminal surface negatively affected by dose reduction in b of polyps 10 mm in diameter or larger and improved differentiation of colorectal polyps from residual stool and hypertrophied folds (with a consequent reduction of false positive diagnoses).

Other authors used a 2.5-beam collimation protocol and demonstrated that, despite a perceptible worsening of image quality, the sensitivity for detection of polyps was equal at 100, 50, and 30 effective mAs (Van Gelder et al. 2002). These results indicated that CTC with MDCT can be reliably performed with an effective dose of 3.6 mSv.

In our experience we used an effective mAs value of 10, demonstrating that MDCT technology can achieve a further, substantial radiation dose reduction for CTC with a total (i.e., prone and supine acquisitions combined) effective dose of 1.8 mSv for men and 2.4 mSv for women. These values are substantially lower than previously published data, both as reported for SSCT and MDCT, and also lower than those of barium enema (5-7 mSv) (Kemerink et al. 2001). This technical approach (defined as "ultra low-dose" technique) was subsequently validated in a large population of patients with results comparable with full dose imaging protocols (Iannaccone et al. 2003b) (Fig. 6.7). Potential criticisms to the use of an ultra low-mAs protocol are as follows: (1) imaging of obese patients, which might be unfeasible although this issue has not been well evaluated yet; (2) poor assessment of low contrast structures, such as liver, pancreas, kidneys, and lymph nodes (Fig. 6.8). This can be expected because the quality of low contrast structures is affected more by noise than that of high contrast structures (i.e., colonic mucosa-air interface). Thus, this imaging protocol will prevent the detection of extra-colonic findings unless deionising filters are implemented.

Another feasible way to reduce patient dose delivery might be a combined approach of ultra low-dose protocol for prone scan and normal tube current setting for supine scan in order to achieve reduction of patient exposure, having at the same time a consistent evaluation of extra-colonic findings. This might be an optimal solution for symptomatic patients, especially if contrast medium injection is required. In a non-symptomatic screening population, both the acquisitions might be achieved using low dose protocols (Nicholson et al. 2005).

A recent technological advancement is represented by the development of automatic dose delivery system, able to modulate the tube current on the basis of the depth of tissues to be scanned. In other words, this technique allows one to reduce patient radiation dose modifying the tube output according to the patient geometry during each rotation and in the longitudinal direction.

Another technical parameter influencing dose delivery is tube potential. Tube potential is expressed as kilovolt peak (kVp). Modification of kVp leads to changes of the photon beam energy expressed as kiloelectron volt (keV). Increasing tube potential, photon beam are more penetrating resulting in an enhancement of detector energy fluency. Modification of tube potential leads to changes in image noise, contrast resolution and patient dose exposure. The most important effect of kVp changes is represented by increase of patient radiation exposure. CTDI increase linearly with mAs and exponentially with kVp (©kVp2). Increase in tube potential b a

Fig. 6.8. Low dose scan of liver parenchyma. Image noise prevents the assessment of focal liver lesions and makes difficult even the visualization of the gallbladder

Fig. 6.7a-c. Low dose scan of carcinomatous polyp of the right colon. Although: a noise clearly degrades image quality, the evaluation using: b wide window centre and setting as well as: c three-dimensional endoluminal view are not significantly affected produces decrease of image noise but also decrease of HU values of different structures due to a stronger detector energy fluency. Decrease of HU values using high values of tube potential is more pronounced for structures with intrinsic high Z, like bone or iodine rather than fat or muscles. Thus increase of tube potential leads to reduction of contrast-resolution for high density materials (Huda et al. 2000). Due to huge impact on patient radiation exposure, increase of kVp in order to reduce image noise should not be used.

Practical Guidelines

A single scanning protocol with identical parameters for all scanners and patients cannot be recom-

mended due to technological differences as well as different clinical indications to CTC.

General guidelines are provided by the recently published Consensus Statement (Barish et al. 2005). If considering basic parameters (i.e., col-limation and mAs values) recommendations propose a collimation no larger than 5 mm for SSCT and no larger than 3 mm for MDCT. With the advent of 64-slice MDCT, sub-millimeter collima-tion will be mandatory, although clinical benefits are still unclear.

Image reconstruction should be overlapped with SSCT: it is usually set at 3 mm for 5-mm collimation protocols and at 1 mm for 3-mm collimation protocols (Sanjay 2004). On MDCT, when using 3 mm effective slice thickness, images are usually reconstructed at 1 mm; when 1 mm or sub-millimeter effective slice thicknesses are used, 1 mm might be the best compromise considering also the number of images to be managed on the workstation.

Considering dose exposure, if CTC is performed as a screening method in asymptomatic population, tube current must be set at the minimum level possible that allows adequate visualization of the colonic wall even if visualization of parenchymatous organs is reduced. There is no consensus at the moment regarding the value of detecting the extra-colonic findings.

If patient is symptomatic or he/she requires i.v. injection of contrast medium for any other clinical reason, mAs should be set at the standard value for an abdomen and pelvis CT scan. In order to reduce dose delivery, patient might be scanned in prone position using a low or an ultra-low dose protocol, and at full dose only in supine scan when contrast medium is injected.

Appendix. Glossary of Terms

Adapted from Flohr et al. (2005) Automatic exposure control

This is a recently developed technique that allows one to reduce patient radiation dose, modifying tube output according to the patient geometry during each rotation and in the longitudinal direction.

Collimated section thickness

Active width of one detector row in the longitudinal direction, measured at the isocenter. The collimated section thickness can be different from the effective section thickness established in the spiral reconstruction process.

Cone angle

With multi-detector row CT system, the measurement rays are no longer perpendicular to the longitudinal axis but are tilted by the cone angle with respect to the center plane. The cone angle is largest for the sections at the outer edges of the detector.

Weighted CT dose index (CTDIw)

Dose measure that uses the absorbed dose in an acrylic plastic phantom. Weighted CTDI (CTDIw) is approximation to average dose in the x-y plane for phantom and so is an approximation of the dose delivered to a cross section of the patient's anatomy. CTDIw is measured in milligrays. CTDIw does not consider exposure variation along the z-axis.

CTDIvol

CTDIvol takes account of exposure variation in the z axis and is calculated by the following formula: CTDIvol=CTDIw Pitch

Deionising filters

Filters implemented from some vendors in the processing of raw data with the aim of reduce image noise.

Dose length product: is measure of total radiation exposure for the whole series of image and is calculated by the following formula: DLP=CTDIvol scan length. Usually in helical CT irradiated length is longer than imaged length. CTDIvol is independent of scan length. DLP is proportional to scan length.

Effective patient dose

Approximation of the dose delivered to the patient during a CT scan that takes into account the different organ sensitivities to radiation. Effective patient dose is measured in millisieverts.

Effective section thickness

Section thickness that is established in the spiral reconstruction process, measured at the isocenter. It is equal to or larger than the collimated section width.

Hyperplane

Intermediate image plane used in a reconstruction method for multi-detector row spiral CT.

Using MDCT, X-ray beam is no longer perpendicular to detector. To overcome this problem vendors have developed different reconstruction algorithms; some of them use an intermediate image plane to reconstruct the final axial image. This algorithm is represented by the so-called hyperplane weighted reconstruction.

Isocenter

Centre of the measurement field of view of a CT scanner. Physical parameters such as section thickness and resolution are usually determined at the isocenter.

Longitudinal resolution

Spatial resolution in the patient (z-axis) direction, determined by the effective section thickness and the image increment.

Pitch

Parameter that characterizes a spiral scan, defined as table feed per rotation divided by the total width of the collimated beam. In the early days of four-section CT, the term detector pitch had been additionally introduced, which accounts for the width of a single section in the denominator. In this way two different pitch can be defined relative to total X-ray collimation (pitchx) or to individual detector width (pitchd). Consequently pitchd==pitchx number of slices. For the sake of clarity and uniformity, the detector pitch should no longer be used. For both single-section CT and multi-detector row CT the pitch (p) is given by p=TF/w, where TF is table feed per rotation, and w is total width of the collimated beam, according to International Elec-trotechnical Commission specifications. For p<1, data acquisition occurs with overlap in the z-axis direction; for p>1, data acquisition occurs with gaps.

Prepatient collimation

Collimation system between X-ray tube and patient that limits the width of the collimated beam in the longitudinal direction.

Section sensitivity profile (SSP)

Function indicating the signal contribution of an infinitesimal object to each position along the z axis. SSP is determined by section collimation, size of the focal spot, and the spiral reconstruction algorithm.

Tube current

Acquisition parameter expressed in terms of milliAmpere (mA) strictly related to image noise and patient dose exposure.

Tube potential

Acquisition parameter expressed in terms of kilovolts peak strictly related to image noise and patient dose exposure (kVp).

Voxel

Basic 3D element of a volume image. A value, the CT number, is assigned to each voxel.

Volume-rendering technique (VRT) Three-dimensional postprocessing method based on a weighted display of all voxels along each ray in the view direction. Transfer functions assign opacity and color to each CT number.

Weighted hyperplane reconstruction (WHR)

Reconstruction method for multi-detector row spiral CT that accounts for the cone angle of the measurement rays by introducing intermediate hyperplanes for image reconstruction.

Z-filter reconstruction

Method for multi-detector row spiral interpolation that uses all direct and complementary rays within a selectable distance from the image plane. Z-filter reconstruction allows establishment of different SSPs to trade off longitudinal resolution and image noise.

References

Barish MA, Soto AJ, Ferrucci JT (2005) Consensus on current clinical practice of virtual colonoscopy Am J Roentgenol 184:786-792

Beaulieu CF, Napel S, Daniel BL et al. (1998) Detection of colonic polyps in a phantom model: implications for virtual colonoscopy data acquisition. J Comput Assist Tomogr 22:656-663

Berland LL, Smith JK (1998) Multidetector-array CT: once again, technology creates new opportunities. Radiology 209:327

Calhoun PS, Kuszyk BS et al. (1999) Three-dimensional volume rendering of spiral CT data: theory and method. RadioGraphics 19:745

Chen SC, Lu DSK, Hecht JR et al. (1999) CT colonography: value of scanning in both the supine and prone positions. Am J Roentgenol 172: 595-599

Dachman AH, Lieberman J, Osnis RB et al. (1997) Small simulated polyps in pig colon: sensitivity of CT virtual cologra-phy. Radiology 203:427-430

Fenlon HM, Nunes DP, Schroy PC III, Barish MA et al. (1999) A comparison of virtual and conventional colonoscopy for the detection of colorectal polyps. N Engl J Med 341:14961503

Fletcher JG, Johnson CD, Welch TJ et al. (2000) Optimization of CT colonography technique: prospective trial in 180 patients. Radiology 216:704-711 Flohr TG, Schaller S, Stierstorfer K et al. (2005) Multi-dete-cor row CT systems and image-reconstruction techniques. Radiology 235:756-773 Giacomuzzi SM, Torbica P, Rieger M et al. (2001) Radiation exposure in single slice and multi-slice spiral CT (a phantom study). Fortschr Rontgenstr (ROFO) 173(7):643-649

Gillams AR, Lees WR (2002) What collimation is necessary for polyp detection at VC? Proceedings of the 3rd International Workshop on Multislice CT, 3D Imaging, Virtual Endoscopy, June 5-8, 2002, Rome, Italy Hara AK, Johnson CD, Reed JE et al. (1997a) Reducing data size and radiation dose for CT colonography. Am J Roentgenol 168:1181-1184

Hara AK, Johnson CD, Reed JE et al. (1997b) Detection of colorectal polyps with CT colography: initial assessment of sensitivity and specificity. Radiology 205:59-65 Hara AK, Johnson CD, MacCarty RL et al. (2001) CT colonography: single-versus multi-detector row imaging. Radiology 219:461-465 Hu H (1999) Multi-slice helical CT: scan and reconstruction.

Med Phys 26:5-18 Hu H, He HD, Foley WD et al. (2000) Four multidetector-row helical CT: image quality and volume coverage speed. Radiology 215:55-62 Huda W, Scalzetti EM, Galina L et al. (2000) Technique factors and image quality as functions of patient weight at abdominal CT. Radiology 217:430-435 Iannaccone R, Laghi A, Catalano C et al. (2003a) Feasibility of ultra-low-dose multislice CT colonography for the detection of colorectal lesions: preliminary experience. Eur Radiol 13(6):1297-1302 Iannaccone R, Laghi A, Catalano C et al. (2003b) Performance of lower dose multi-detector row helical CT colonography compared with conventional colonoscopy in the detection of colorectal lesions. Radiology 229(3):775-781 Kemerink GJ, Bortslap AC, Frantzen MJ et al. (2001) Patients and occupational dosimetry in double contrast barium enema examinations Br J Radiol 74(881):420-428 Laghi A, Iannaccone R, Mangiapane F et al. (2003) Experimental colonic phantom for the evaluation of the optimal scanning technique for CT colonography using a multidetector spiral CT equipment. Eur Radiol 13:459-466 Macari Ml, Bini EJ, Xue X et al. (2002) Colorectal neoplasms:

prospective comparison of thin-section low-dose multi-detector row CT colonography and conventional colonos-copy for detection. Radiology 224:383-392 McCollough CH (2002) Optimization of multidetector array CT acquisition parameters for CT colonography. Abdom Imaging 27:253-259 Nicholson FB, Taylor S, Halligan S et al. (2005) Recent developments in CT colonography. Clin Radiol 60: 1-7 Power NP, Pryor AM (2002) Optimization of scanning parameters for CT colonography. Br J Radiol 75:401-408 Rogalla P, Meiri N (2001) CT colonography: data acquisition and patient preparation techniques. Semin Ultrasound CT MR 22:405-412

Rottgen R, Schroder RJ, Lorenz M et al. (2003) CT-colonog-raphy with the 16-slice CT for the diagnostic evaluation of colorectal neoplasms and inflammatory colon diseases. Fortschr Rontgenstr (ROFO) 175:1384-1391 Sanjay S (2004) Multi-detector row CT: principles and practice for abdominal applications. Radiology 233:323-327 Sherbondy AJ, Holmlund D, Rubin GD et al. (2005) Alternative input devices for efficient navigation of large CT angiography data sets. Radiology 234(2):391-398 Springer P, Stohr B, Giacomuzzi SM et al. (2000) Virtual computed tomography colonoscopy: artefacts, image quality and radiation dose load in a cadaver study. Eur Radiol 10:183-187

Taguchi K, Aradate H (1998) Algorithm for image reconstruction in multi-slice helical CT. Med Phys 25:550-561 Taylor SA, Halligan S, Bartram I et al. (2003) Multi-detector row CT colonography: effect of collimation, pitch, and orientation on polyp detection in a human colectomy specimen. Radiology 229:109-118 Van Gelder RE, Venema HW, Serlie IW et al. (2002) CT colo-nography at different radiation dose levels: feasibility of dose reduction. Radiology 224:25-33 Vining DJ, Gelfand DW, Bechtold RE et al. (1994) Technical feasibility of colon imaging with helical CT and virtual reality. Am J Roentgenol 62 Suppl:104 Wessling J, Fischbach R, Meier N et al. (2003) CT colonog-raphy: protocol optimization with multi detector row CT - study in an anthropomorphic colon phantom. Radiology 228:753-759

Whiting BR, Mc Farland EG, Brink JA et al. (2000) Influence of image acquisition parameters on CT artefacts and polyp depiction in spiral CT colonography: in vitro evaluation. Radiology 217:165-172

Was this article helpful?

0 0
Managing Diverticular Disease

Managing Diverticular Disease

Stop The Pain. Manage Your Diverticular Disease And Live A Pain Free Life. No Pain, No Fear, Full Control Normal Life Again. Diverticular Disease can stop you from doing all the things you love. Seeing friends, playing with the kids... even trying to watch your favorite television shows.

Get My Free Ebook


Post a comment