WO2013159792A1

WO2013159792A1 - Method and device for computed tomography angiography

Info

Publication number: WO2013159792A1
Application number: PCT/EP2012/001798
Authority: WO
Inventors: Annika SCHUHBÄCK; Stephan ACHENBACH
Original assignee: Universitätsklinikum Erlangen
Priority date: 2012-04-26
Filing date: 2012-04-26
Publication date: 2013-10-31

Abstract

A method of computed tomography (CT) imaging, in particular for CT angiography for imaging vessels in a region of interest (ROI) of a proband body, includes the steps of irradiating (S1) the proband body along a plurality of projection directions with X-rays created with an X-ray tube of a CT imaging device, wherein an energy input provided by the X-rays is adjusted (S2) in dependency of a proband-specific parameter, collecting (S3) a plurality of projection data of the proband body, and reconstructing (S4) at least one tomographic image of the ROI on the basis of the projection data, wherein the proband-specific parameter is a radial size quantity of a body section surrounding the ROI. Furthermore, a computed tomography (CT) imaging device, in particular adapted for implementing the above method, and a method of using the radial size quantity of the body section surrounding the ROI as a predictor of image noise in CT imaging are described.

Description

Method and device for computed tomography angiography Field of the invention

The invention relates to a method of computed tomography (CT) imaging, in particular to a method of CT angiography for imaging vessels in a region of interest (ROI) of a proband body, like e. g. coronary CT angiography. Furthermore, the invention relates to a CT imaging device, in particular being adapted for CT angiography. Applications of the invention are present in the fields of medical CT imaging for diagnostic or prophylactic purposes.

Background of the invention

Coronary computed tomography (CT) angiography permits the detection of coronary artery stenoses with high sensitivity and specificity (see S. Achenbach in "J. Cardiovasc. Comput . To- mogr." 2007;1:3-20). The main indication for coronary CT angiography is to rule out significant coronary artery stenoses in symptomatic patients with an intermediate likelihood for coronary artery disease. The future event rate in symptomatic patients with a "negative" coronary CT angiogram is extremely low. Coronary CT angiography can be performed with high patient comfort and a low rate of acute complications. However, it can be associated with high radiation exposure (see J. Hausleiter et al. in "JAMA" 2009; 301 : 500- 507; and J. Haus- leiter et al. in "Circulation" 2006;113:1305-1310). On the other hand, new image acquisition protocols enable the performance of coronary CT angiography with substantially reduced radiation exposure. With some acquisition protocols, estimated effective doses of less than 1.0 mSv are achievable in selected patients (see S. Achenbach et al. in "Eur. Heart J." 2010;31:340-346). Among the parameters that can be adjusted to lower energy input (radiation exposure) , reductions in tube current and reductions in tube voltage are often used. For example, reducing tube voltage from 120 kV to 100 kV tube voltage can reduce the dose by up to 40 % (see e. g. J. Hausleiter et al. in "Circulation" 2006;113:1305-1310).

However, reduction of tube current and voltage will both lead to higher image noise (see e. g. S. Leschka et al. "Eur. Radiol." 2008;18:1809-1817). Increased image noise may compromise diagnostic accuracy and therefore, the use of low-dose acquisition protocols needs to be tailored to the individual patient (proband) in order to avoid inacceptable high image noise. Therefore, it has been proposed in the past to adjust tube current and voltage in dependency on a proband-specific parameter .

Several proband-specific parameters such as body weight or body mass index (see N. Yoshimura et al. in "Acad. Radiol." 2006;13:324-328; and J. Horiguchi et al. in "Korean J. Radiol." 2009;10:340-346), proband circumference (see J. Menke et al. in "Radiology" 2005;236:565-571), chest area, chest attenuation (see B. B. Ghoshhajra et al. in "J. Cardiovasc. Comput. Tomogr." 2011;5:240-246), transverse chest diameter (see M. K. Kalra et al. in "Korean J. Radiol." 2003;4:234- 238), and X-ray attenuation in the scout view (see J. Gao et al. in "Eur. J. Radiol." 2011;79:266-271) have been studied so far, and have shown a certain relationship to image noise or image quality. Rogalla et al. (in "Int. J. Cardiovasc Imaging" 2010;26:933-940) demonstrated that the anterioposterior chest diameter assessed from the lateral scout view was an appropriate method to adapt the tube current in coronary CT angiography while image quality could be preserved. Wink- lehner et al. (in "Invest. Radiology" 2011;46:767-773) evaluated an automated attenuation-based tube potential selection based on the attenuation along the proband' s longitudinal axis acquired by topogram images for thoracoabdominal CT an- giography. They could demonstrate that this method reduced overall radiation dose by 25% as compared to standard 120 kV.

However, practical investigations have shown that the conventionally used proband-specific parameters do not allow an ap- plication in medical routine investigations. There is a poor understanding of the relationship between proband characteristics and image noise, and suitable predictors of image noise allowing CT angiography imaging with a sufficient reliability have not yet been identified.

The above problems are related not only to CT coronary angiography, but also to CT angiography of other body portions, like e. g. angiography of the brain. Objective of the invention

The objective of the invention is to provide an improved method of CT imaging, in particular CT angiography, which is capable of avoiding disadvantages of conventional CT imaging. In particular, the objective is to provide an improved CT imaging method which provides a reliable dose reduction while keeping a practical image quality. Furthermore, the objective of the invention is to provide an improved CT imaging device, in particular for CT angiography purposes, which is capable of avoiding disadvantages of conventional CT imaging devices. Summary of the invention

The above objective is solved by a CT imaging method and a CT imaging device, resp., comprising the features of the inde- pendent claims. Advantageous embodiments of the invention are defined in the dependent claims.

According to a first aspect of the invention, a method of CT imaging, in particular for CT angiography for imaging vessels in a region of interest (ROI) of a proband body, is provided, wherein the proband body including the ROI is irradiated along a plurality of angular projection directions in an imaging plane with X-rays created with an X-ray source device. Projection data of the proband body are collected represent- ing attenuation values of the proband body along the projection directions. At least one tomographic image of the ROI is reconstructed on the basis of the projection data. Preferably, the inventive method is applied for medical imaging, i. e. the proband typically is a person (patient) to be sub- jected to a diagnosis.

The energy input provided by the X-rays in the proband body is adjusted in dependency of a proband-specific parameter. Adjusting the energy input comprises minimizing the energy input such that the projection data can be collected and at least one tomographic image can be reconstructed without an image noise which would compromise a diagnostic evaluation of at least one tomographic image. Image noise can be repre^¬ sented with a HU noise value (noise value in Hounsfield

Units) . Preferably, the energy input is set such that at least one tomographic image can be reconstructed with an image noise below a predetermined noise threshold value. The acceptable noise threshold value may depend on the type of diagnosis information to be obtained by the CT angiography, e. g. whether coronary artery stenoses or plaque deposits are to be identified in at least one tomographic image. Alternatively, the acceptable noise threshold value can be set to a value depending on the application conditions, e. g. by a user of the CT imaging device. As a general rule, energy input is set such that the HU noise value is below 40.

According to the invention, the proband-specific parameter used as a measure for adjusting the energy input is a radial size quantity (or: radial dimension) of a body section surrounding the ROI, i. e. an average thickness and/or an area of the body section delimited in radial direction in the imaging plane. The radial directions are provided by a connection line between the X-ray source device and the detector device. The inventors have done extended studies for analyzing the impact of several parameters, like e. g. the area of the thoracic cross section (area of the thoracic solid tissue) , on image noise. The inventors have found that the radial size quantity of the body section adjacent to the ROI and excluding the ROI, like e. g. a chest wall (thorax wall) allows an energy input adjustment with an essentially improved reliability. Further, the inventors have found that the radial size quantity of the body section surrounding the ROI provides a measure which is independent of eventual further specific physiological features of the proband. As an example, a little person with a low body weight or a female patient with a low body weight may have a thick chest wall, in particular due to breast tis- sue. If the energy input is reduced due to the low weight, the thick chest wall may compromise the image quality in the imaging plane. In other words, contrary to conventional approaches, where a global quantity of the proband, like e. g. the weight, circumference or whole sectional area thereof was used as a proband-specific measure, the radial size quantity of the body section surrounding the ROI provides a specific local measure for an overall attenuation of the X-rays in the imaging plane.

Advantageously, the invention allows a dose reduction while keeping the image quality on a reliable diagnostic level. The invention provides a generally usable predictor for the image noise in tomographic images which directly can be used for adjusting a minimized energy input. As an example, reduction of 120 kV to 100 kV tube voltage can reduce the dose by up to 40 %.

According to a second aspect of the invention, a CT imaging device, in particular adapted for CT angiography for imaging vessels in a region of interest (ROI) of a proband body, is provided, which comprises an X-ray source device arranged for irradiating the proband body with X-rays along a plurality of projection directions, an adjustment device arranged for ad- justing an energy input provided by the X-rays in dependency of a proband-specific parameter, a detector device arranged collecting a plurality of projection data of the proband body, and an image reconstruction device being arranged for reconstructing at least one tomographic image of the ROI on the basis of the projection data. According to the invention, the adjustment device is configured for adjusting the energy input in dependency of a radial size quantity of a body section surrounding the ROI. Preferably, the CT imaging device is adapted for conducting the method of CT imaging according to the above first aspect of the invention.

According to a third aspect of the invention, a method of providing a predictor of image noise in computed tomography (CT) imaging, in particular in CT angiography for imaging vessels in a region of interest (ROI) of a proband body, is provided, which includes the steps of collecting a preview tomographic image of the proband and determining a radial size quantity of a body section surrounding the ROI using the preview tomographic image, wherein the radial size quantity provides the image noise predictor to be obtained. Preferably, the area of the thorax wall (thoracic solid tissue) at the level of the aortic root predicts the image noise and may hence be used for the decision to reduce tube voltage e. g. from 120 kV to 100 kV.

Advantageously, multiple variants of using a radial size quantity of a body section surrounding the ROI exist, which can be selected in dependency on the particular application of the invention. With a first, preferred variant (coronary CT angiography) , the ROI is the heart of the proband, while the body section considered is the proband's chest wall (thorax wall) surrounding the heart. In this case, energy input of the X-rays is adjusted in dependency on the integrated area of the chest wall and/or a thickness of the chest wall surrounding the heart. With a further variant (brain CT angiography) , the area and/or thickness of a cranium surrounding the proband's brain is used as a measure for adjusting the energy input of the X-rays. Furthermore, with abdomen CT angiography, the area and/or thickness of an abdominal wall surrounding the proband' s abdomen is the quantity which controls the adjustment of the energy input.

Accordingly, the adjustment device of the inventive CT imag- ing device is adapted for receiving at least one of the above radial size quantity quantities and/or any data being a function of at least one of the above radial size quantity quantities as an input, e. g. from an external data storage or from another circuitry of the CT imaging device. According to a preferred embodiment of the invention, the energy input to be set is selected by comparing the radial size quantity of the particular body section with reference data which provide a relationship of radial size quantities and image noise depending on the energy input. Thus, the reference data provide a minimum energy input required so that the image noise falls below the predetermined noise threshold value. As an advantage, controlling the energy input can be done quickly and without complex procedural steps. The reference data can be stored in a reference data storage device connected with the adjustment device. Preferably, the reference data comprise discrete values, e. g. a reference table, or continuously distributed values, like e. g. a reference calibration function. The reference data can be obtained from reference measurements with multiple probands and/or from theoretical predictions based on simulations.

The radial size quantity of the body section surrounding the ROI can be input to the adjustment device from an external data storage. As an example, the radial size quantity of the chest wall with multiple imaging planes can be measured at a previous tomographic imaging of the proband and stored with the proband's data for further uses. In this case, the radial size quantity can be obtained without an additional irradiation of the proband.

However, according to a particularly preferred embodiment of the invention, the radial size quantity is obtained from a measurement just before the angiography imaging to be conducted. A preview tomographic image of the proband is collected, and the radial size quantity is calculated using the preview tomographic image. With this embodiment, the CT imaging device includes a calculation device for determining the radial size quantity by processing the preview tomographic image of the proband.

According to a preferred variant of the invention, the pre- view tomographic image can be collected in a contrast agent free condition of the proband. Advantageously, the exposure of the proband to the contrast agent can be reduced. Alternatively, the preview tomographic image can be collected in a contrast-enhanced condition before the CT angiography of the proband ("test bolus" imaging). In both cases, as a particular advantage of the invention, the inventive method does not need to be changed depending on where a contrast-free or contrast-enhanced image is collected. The inventive adjustment of the energy input can be done by a user of the CT imaging device. As an example, the user can collect the preview tomographic image, select the body section surrounding the ROI, calculate the radial size quantity of the body section and manually adjust the energy input in dependency on the calculated radial size quantity. Alternatively, with a further preferred embodiment of the invention, the energy input can be automatically adjusted using the CT imaging device. The inventive CT imaging device may include a control device which is adapted for setting the X-ray source device such that a certain energy input is obtained. According to particularly preferred implementations of the invention, at least one of a tube voltage and a tube current of the at least X-ray tube of the X-ray source device is controlled automatically. Brief description of the drawings

Further details and advantages of the invention are described in the following with reference to the attached drawings, which show in:

Figure 1: a flow chart of a CT angiography method according to a preferred embodiment of the invention;

Figures 2 to 6: flow charts and practical examples illustrat ing further details of the CT angiography method shown in Figure 1; and

Figure 7: a schematic illustration of a CT imaging device ac- cording to a preferred embodiment of the invention.

Description of preferred embodiments of the invention

Preferred embodiments of the invention are described in the following with particular reference to the CT angiography dose reduction in dependency on proband-specific parameters, like the thorax wall area. Details of CT imaging methods and devices and image diagnostics are not described as far as they are known from conventional CT techniques. In particu- lar, it is emphasized that the implementation of the invention is not restricted to a certain CT scanner geometry, CT data collection method or CT image reconstruction method. As an example, the invention is described below with reference to a CT imaging device, wherein the X-ray source device is a single X-ray tube rotatable on the scanner gantry. With other embodiments of the invention, the X-ray source device may comprise multiple X-ray tubes arranged around the proband body. Furthermore, the features of preferred embodiments are described in the following with exemplary reference to coro- nary CT angiography. The invention can be correspondingly used with other angiography applications, like brain or abdomen angiography. 1. Method of computed tomography imaging

Figure 1 schematically illustrates the main steps of a preferred embodiment of a coronary CT angiography method according to the invention. This CT angiography method is imple- mented with a CT imaging device, e. g. as shown in Figure 7, comprising one movable X-ray tube emitting X-rays, a detector device collecting a plurality of projection data of the proband body, and an image reconstruction device. The heart of a proband provides the ROI to be imaged. Tomographic images of the heart, in particular coronary vessels, are to be subjected to a diagnosis of eventual stenoses or plaque depositions. The body section surrounding the ROI is the thorax wall . In a first step SI, a proband-specific radial size quantity of the thorax wall is provided. As an example, the area of the thorax wall is input from an external data storage or by a user of the CT imaging device, or it is determined with the CT imaging device as described with further details below (see Figures 2 and 3) .

Subsequently, the X-ray tube of the CT device is adjusted such that a predetermined energy input of the X-rays is provided in the ROI. The goal of the energy input adjustment is to set the tube voltage and/or the tube current of the X-ray tube such that the tomographic image of the heart has an image noise below a predetermined noise level, e. g. below 40 HU units. To this end, the redial size quantity provided in step SI is compared with reference data as described with further details below (Figures 4 and 5) .

In the following, the steps of irradiating the proband body along a plurality of projection directions with X-rays and collecting a plurality of projection data of the proband body (S4) as well as reconstructing of at least one tomographic image of the ROI on the basis of the projection data (S5) are conducted. Steps S4 and S5 are implemented as known from con- ventional CT imaging techniques in dependency on the particular CT scanner type practically used.

According to Figure 2, step SI of providing the radial size quantity preferably comprises the sub-steps of reconstructing a preview tomographic image of the proband body in the imaging plane including the ROI (Sll) and determining the radial size quantity on the basis of the preview tomographic image (S12) . Thus, Figure 2 illustrates an embodiment of a method of providing a predictor of image noise in CT imaging. Pref- erably, the preview tomographic image used for providing the radial size quantity for coronary CT angiography is selected to intersect the aortic root of the heart, in particular at the level of the left main coronary artery. As an example, the preview tomographic image is collected with a contrast agent free condition of the proband. As an example, the preview tomographic image is a Ca-score-image which can be collected with a low tube current and low radiation dose. Other examples are transaxial scout images, or from images acquired for the purpose of determining the contrast agent transit time. Figures 3A and 3B show examples of the preview tomographic image including sectional views of the proband' s 1 thorax with high attenuation along the thorax wall 2 and in the heart 3. The lung 4 surrounded by the tho- rax wall 2 has a lower attenuation. The radial size quantity, like e. g. the area of the thorax wall 2 (hatched in Figure 3A) is determined in step S12 (Figure 2) by subjecting the preview tomographic image to an image analyses procedure. This is an important distinction over conventional techniques where the area of the complete thorax 6 (hatched in Figure 3B) is used as a measure for the image noise in CT imaging. As an example, the body section representing the thorax wall 2 (Figure 3A) is automatically recognized by a threshold value analyses of the attenuation in the preview tomographic image and a shape detection. Subsequently, the area of the recognized thorax wall 2 is calculated. As an alternative to the automatic thorax wall recognition, the user can mark the thorax wall in the preview tomographic image, and the thorax wall area is calculated for the marked section.

As an alternative to the thorax wall area as shown in Figure 3A, a thickness of the thorax wall can be used for obtaining the radial size quantity. As an example, the thickness along at least one certain radial line (e. g. at 5) is calculated using the preview tomographic image. If the radial size quantity is based on the thickness of the thorax wall, an average thickness can be calculated using thicknesses measured along different radial projection directions through the thorax wall.

According to Figure 4, step S2 of adjusting the X-ray tube includes the sub-steps of comparing the radial size quantity with predetermined reference data (S21) and setting a minimum energy input of the X-ray tube (S22) .

The reference data comprise calibration curves as schematically shown in Figure 5. The calibration curve show the energy input (tube voltage) to be selected in dependency on the area of the thorax wall for certain tube currents. With an increasing area, an increasing tube voltage is set. The calibration curves are obtained from correlations measurements (see below, Figure 6) . Further calibration curves can be stored presenting tube currents to be selected in dependency on the area of the thoracic wall for certain tube voltages.

The comparison of the thoracic wall area with the reference data can be done by a user of the CT imaging device or auto- matically by the control circuit included in the CT imaging device. Subsequently, the energy input of the X-ray tube is set depending on the thoracic wall area and the acceptable image noise. The calibration curves of Figure 5 can be obtained from correlation measurements as follows. Figure 6 shows an example of a correlation of the image noise (HU) with the thoracic wall area for a certain tube voltage of the X-ray tube (e. g. 100 kV) . As an example, if the proband has a thoracic wall area of 500 cm² and an image noise of 40 HU would be acceptable, the correlation curve shows that the tube voltage should be adjusted to a higher value, e. g. about 120 kV representing the corresponding minimum energy input value in Figure 5. Complete calibration curves are based on multiple correlation curves collected with different X-ray tube voltages and currents.

2. CT imaging device Figure 7 schematically illustrates a preferred embodiment of an CT imaging device 100 comprising an X-ray source device 10, a detector device 20, an image reconstruction device 30, a holding device 40, a control device 50 including an adjustment device 51, and an output device 60 with e. g. a display device, a data recording device and/or an image printer. Basically, the components 10, 20, 40 and 60 are structured as it is known from conventional CT scanners, like e. g. the Dual source CT scanner (Definition Flash, Siemens Healthcare, Forchheim, Germany) . In particular, the X-ray source device 10 is an X-ray tube which is rotatable on the CT scanner gantry 101 around the proband 1 on the holding device 40. The operation of the X-ray tube can be adjusted by controlling the tube voltage and the tube current.

The image reconstruction device 30 includes a projection data processing unit 31, an image reconstruction unit 32 and a calculation device 33. The projection data processing unit 31 and the image reconstruction unit 32 operate as it is known from conventional CT scanner techniques. The calculation device 33 is a circuitry which calculates the radial size quantity to be obtained, like e. g. the thoracic wall area from the image data provided by the image reconstruction unit 32. The control device 50 includes an input circuit 53 receiving the radial size quantity calculated with the calculation device 33, a reference data storage 54, a comparing device 52 and the adjustment device 51. Additionally, a general control unit 55 can be provided as it is known from conventional CT scanners.

With the CT imaging device 100, the inventive CT imaging method is conducted as described above. In particular, the CT imaging device 100 is controlled to collect the preview tomographic image for calculating the thorax wall area. After comparing the calculated thoracic wall area with the reference data, the tube voltage and/or the tube current are adjusted with the adjustment device 51 such that the image noise is below a predetermined noise threshold. Subsequently, the coronary angiography image is collected. In the following, the angiography image is subjected to further image analyses for diagnostic purposes as it is known from conventional angiography techniques.

3. Experimental results

The invention is based on inventor's findings, which have been derived from experiments summarized in the following. Contrast-enhanced coronary dual source CT angiography data sets (Dual Source CT Scanner Definition Flash, 280 ms rotation, 2x128x0.6 mm collimation, in deep inspiration, prospectively ECG-triggered axial and retrospectively ECG-gated spiral acquisition, 100 kV, 320 mAs) of 165 probands (age 54+13 years) for the detection of coronary artery stenoses were analyzed. Image noise was measured in the aortic root. Influence of body weight, height, body mass index, thoracic cross sectional area as well as the area of the thoracic solid tissue were analyzed.

Probands with a heart rate >65 beats per minute (bpm) received 50 or 100 mg atenolol orally at least 30 minutes before the CT scan. If heart rate remained >65 bpm in inspiration, up to 30 mg metoprolol was injected intravenously, using repeated 5-mg doses before CT. Prior to coronary CT angiography, all probands received 0.8 mg glycerol trinitrate sublingually. A "test bolus" protocol was used. 10 ml of contrast agent (Iomeprol, Imeron 350, Bracco Altana Pharma GmbH, Germany) were injected at a flow rate of 5 ml/s, followed by 50 ml of saline at the same flow rate. The time to peak enhancement in the ascending aorta was measured using a series of axial scans acquired in 2-s increments, with the first image being acquired 15 s after the start of injection. For the coronary CT angiography, 60 ml of contrast agent were in- jected, followed by a 60 ml flush (consisting of 80% saline and 20% contrast), both at the same flow rate of 6 ml/s. Image acquisition was started with a delay corresponding to the contrast agent transit time plus 2 s. Coronary CT angiography data sets were acquired using either prospectively ECG- triggered axial or retrospective ECG-gated spiral acquisition depending on heart rate.

CT angiography images were reconstructed with 0.6 mm slice thickness and an increment of 0.3 mm using a medium smooth reconstruction kernel ("B26f") . In prospectively ECG- triggered axial acquisition, only one time instant in the cardiac cycle was available for image reconstruction (70% of the RR interval) . In retrospectively ECG-gated spiral acquisition, an automatic algorithm detected the optimal phase for image reconstruction in diastole or systole (mean 68% of the RR interval) . CT images for the measurement of the area of the thoracic cross section and the thoracic solid tissue were reconstructed from the same raw data with a full field-of- view of the entire thorax with 3 mm slice thickness and an increment of 1.5 mm using a very sharp reconstruction kernel ("B70f") .

The area of the thoracic cross section and the area of the thoracic solid tissue were measured at the level of the ascending aorta at the origin of the left main coronary artery (see Figure 3A) . The thoracic solid tissue (thorax wall) was defined as the area of the thoracic cross section minus the lung area.

Image noise was measured using the standard deviation (SD) of CT attenuation values in a circular region of interest (3.5 cm²) set in the ascending aorta at the level of the left main coronary artery in the coronary CT angiography data set. The effective radiation dose was derived from the dose length product (DLP) and a conversion factor of 0.014 for chest CT in adults according to Bongartz et al. (European Guidelines for Multislice Computed Tomography: Appendix C Funded by the European Commission, 2004). Mean image noise in the aorta was 35.1+8.9 HU. Mean dose length product was 207+184 cm*cGy with an average effective dose of 2.9+2.6 mSv.

For statistical analyses, data were expressed as mean ± SD and ranges for continuous variables. The whole proband cohort was divided into tertiles according to image noise. Body weight, height, body mass index, the area of the thoracic cross section, the area of the thoracic solid tissue as well as the corresponding mean CT densities of the tertiles were compared to each other. A Mann Whitney U-test was used to test for statistical significance. P values < 0.05 were considered to be statistically significant. Multivariable regression analysis was performed including all parameters which had a significant difference between the tertiles of image noise in univariable analysis.

The proband cohort was divided into tertiles according to image noise. Numerous parameters, including BMI and body weight, were significantly different between the highest and lowest fertile. Multivariable regression analysis included body weight, height, BMI, the area of the thoracic cross section and the area of thoracic solid tissue (see Table 3) . The area of thoracic solid tissue was the only independent predictor of image noise (P < 0.0001, R = 0.067, 95% CI: 0.055- 0.079) .

In agreement with previous studies, the inventors were able to show that numerous proband-specific parameters have an impact on image noise in coronary CT angiography. Body mass in- dex has been confirmed as predictor of image noise, but is conceivably of limited value to predict the noise in coronary CT angiography data sets, since the latter is mainly influ^¬ enced by the amount and attenuation of body tissue specifi- cally in the chest and not remote parts of the body.

The features of the invention disclosed in the above descrip^¬ tion, the drawings and the claims can be of significance both individually as well as in combination for the realization of the invention in its various embodiments.

Claims

Claims 1. Method of computed tomography (CT) imaging, in particular for CT angiography for imaging vessels in a region of interest (RQI) of a proband body, including the steps of

- irradiating (SI) the proband body along a plurality of projection directions with X-rays created with an X-ray tube (10) of a CT imaging device (100), wherein an energy input provided by the X-rays is adjusted in dependency of a pro- band-specific parameter (S2) ,

- collecting a plurality of projection data of the proband body (S3) , and

- reconstructing at least one tomographic image of the ROI on the basis of the projection data (S4),

characterized in that

- the proband-specific parameter is a radial size quantity of a body section surrounding the ROI.

2. Method according to claim 1, wherein the radial size quantity is obtained from

- at least one of an area and a thickness of a thorax wall (2) surrounding the proband's (1) heart (3),

- at least one of an area and a thickness of a cranium surrounding the proband' s brain, or

- at least one of an area and a thickness of an abdominal wall surrounding the proband's abdomen.

3. Method according to one of the foregoing claims, wherein

- the energy input to be adjusted is selected by comparing the radial size quantity with reference data providing minimum energy input data required for obtaining a predetermined noise level in the tomographic image to be reconstructed (S21) .

4. Method according to claim 3, wherein

- the reference data comprise at least one of a reference table and a reference calibration function.

5. Method according to one of the foregoing claims, wherein

- the energy input is adjusted such that the at least one tomographic image can be reconstructed with an image noise below a predetermined noise threshold.

6. Method according to claim 5, wherein

- the predetermined noise threshold has an HU noise value below 40.

7. Method according to claim 5 or 6, including at least one of the steps of

- setting the predetermined noise threshold in dependency of the type of diagnostic information to be obtained from the at least one tomographic image, and

- setting the predetermined noise threshold to an application dependent value.

8. Method according to one of the foregoing claims, including the steps of

- collecting a preview tomographic image of the proband

(Sll), and

- determining the radial size quantity of the body section using the preview tomographic image (S12) .

9. Method according to claim 8, wherein

- the preview tomographic image is collected in a contrast agent free condition of the proband.

10. Method according to claim 8, wherein

- the preview tomographic image is collected in a contrast- enhanced condition before the CT angiography of the proband.

11. Method according to claim 9 or 10, wherein

- the CT angiography is conducted without changes depending on where a contrast-free or contrast-enhanced image is collected.

12. Method according to one of the foregoing claims, wherein

- the energy input is automatically adjusted by a control device of the CT imaging device.

13. Method according to one of the foregoing claims, wherein the adjusting step includes at least one of

- setting a tube voltage of the X-ray tube, and

- setting a tube current of the X-ray tube.

14. Computed tomography (CT) imaging device (100), in particular adapted for CT angiography for imaging vessels in a region of interest (ROI) of a proband body (1), including

- a X-ray source device (10) being arranged for irradiating the proband body (1) with X-rays along a plurality of projection directions,

- an adjustment device (51) being arranged for adjusting an energy input provided by the X-rays in dependency of a pro- band-specific parameter,

- a detector device (20) being arranged collecting a plurality of projection data of the proband body (1), and

- an image reconstruction device (30) being arranged for reconstructing at least one tomographic image of the ROI on the basis of the projection data,

characterized in that - the adjustment device (51) is configured for adjusting the energy input in dependency of a radial size quantity of a body section surrounding the ROI .

15. CT imaging device according to claim 14, wherein the adjustment device is arranged for an input of

- at least one of an area and a thickness of a thorax wall (2) surrounding the proband's heart (3),

- at least one of an area and a thickness of a cranium sur- rounding the proband's brain, or

16. CT imaging device according to one of the claims 14 to 15, further including

- a reference data storage device (54) being arranged for storing reference data providing minimum energy input data required for obtaining a predetermined noise level in the tomographic image to be reconstructed, wherein

- the adjustment device (51) is connected with the reference data storage and is capable of selecting the energy input to be adjusted by comparing the radial size quantity with the reference data.

17. CT imaging device according to claim 16, wherein

- the reference data storage device (54) is arranged for storing at least one of a reference table and a reference calibration function.

18. CT imaging device according to one of the claims 14 to

17, including

- a calculation device (33) being arranged for determining the radial size quantity of the body section based on a tomographic image of the proband.

19. CT imaging device according to one of the claims 14 to

18, including

- a control device (50) being arranged for automatically adjusting the energy input.

20. CT imaging device according to one of the claims 14 to

19, wherein the adjustment device (51) is configured for at least one of

- setting a tube voltage of the X-ray source device (10), and - setting a tube current of the X-ray source device (10) .

21. Method of providing a predictor of image noise in computed tomography (CT) imaging, in particular in CT angiography for imaging vessels in a region of interest (ROI) of a proband body, including the steps of

- collecting a preview tomographic image of the proband (1), and

- determining a radial size quantity of a body section surrounding the ROI using the preview tomographic image, wherein - the geometric quantity provides the image noise predictor to be obtained.

22. Method according to claim 21, wherein the radial size quantity is

- at least one of an area and a thickness of a cranium surrounding the proband's brain, or