WO2014070031A1

WO2014070031A1 - Orthogonal computed tomography system

Info

Publication number: WO2014070031A1
Application number: PCT/PT2012/000054
Authority: WO
Inventors: Paulo Crespo; Hugo SIMÕES
Original assignee: Universidade De Coimbra
Priority date: 2011-12-21
Filing date: 2012-12-21
Publication date: 2014-05-08

Abstract

An orthogonal computed tomography (OrthoCT) system for 3D imaging that provides target (e.g. patient) morphological information is presented. The system comprises one or more photon detectors and one or more photon sources. Incoming photon fluxes impinge on the target from opposite directions, with one or several detectors positioned at approximately 90 degrees in respect to the beam axes. We show by simulation, confirmed by experiments, that OrthoCT is capable of providing 3D target density information. Density ratios equivalent or better than those attainable by state-of-the-art tomographic imaging are shown. The simulations also show that these results are obtained with a reduced dose in respect to that delivered by state-of-the-art clinical computed tomography (CT).

Description

ORTHOGONAL COMPUTED TOMOGRAPHY SYSTEM

1 Introduction

The advent of computed tomography (CT) has revolutionized diagnostic radiology (Brenner and Hall 2007). In addition to clinical diagnostic and screening, CT provides, together with modern nuclear medicine imaging technologies such as positron emission tomography (PET) and magnetic resonance imaging (MRI), the basis for radiotherapy treatment planning. To date, all forms of external radiotherapy (RT) treatment planning rely on the information provided by dedicated CT scanners. This is valid for conventional RT techniques such as 3D-conformal RT, as well as for modern, highly-conformal RT techniques such as stereotactic radiosurgery, intensity modulated RT (IMRT), volumetric modulated RT (VMRT) - static and/or dynamic, and helical tomotherapy (HT) , among other. Finally, recent and ongoing developments in the RT arena point towards the field of image-guided radiotherapy (IGRT, e.g. Nath et al. 2009), including the challenging concepts of adaptive radiotherapy (ART) and treatment dose verification.

In either of its clinical applicability diagnostics, screening, and treatment guidance - protection of the patient is of key importance in minimizing carcinogenic and other risks both to individual patients and to the population at large.

This hinders a wider spread of CT usage in clinical practice. In fact, ongoing debates and studies seem to suggest that the clinical benefit provided by CT information may be at times compromised by its associated non-negligible dosage (Brenner and Hall 2007; Nath et al. 2009) . The same debate exists in non-CT diagnostic/screening fields such as mammography, where dose-associated secondary malignancies may also be induced (Andersson and Janzon 1997) . OrthoCT may be equally applied in these fields due to its very high contrast capability even at a very low dose, below 1 mGy.

1.1 CT usage in clinical diagnostics and screening

X-ray imaging constitutes the largest source of dose to the population at large due to artificial ionizing radiation (Pedroso de Lima 2009) . Of these, CT applications are categorized by population of patients (adult or pediatric) and by purpose of imaging (for diagnosis of symptomatic, or screening of asymptomatic patients) . Most CT applied in adults is for the purpose of diagnosis. However, its use in pediatric diagnosis and adult screening has been increasing and can be expected to continue for the next few years even if it is known that CT scans do increase the lifetime risk of cancer, most significantly for pediatric patients (Brenner and Hall 2007) .

In this scenario, where the CT benefit of clinical diagnostics or screening counterparts with its carcinogenic risk at some level, reducing imaging dose would greatly benefit the clinician, the patient, and the population at large. Consequently, a tomographic imaging technique such as OrthoCT represents an undeniable clinical and commercial added value to society due to its intrinsic very low dose capability with no cost for image quality, the latter being even potentially improved as shown here.

1.2 CT usage in the radiotherapy environment

Radiotherapy (RT) plays a well-established role in the management of a wide range of cancers (Alves et al. 2008). Over the last years, many new imaging techniques have been integrated into the RT planning and delivery process. These technologies belong to the category of image-guided radiotherapy (IGRT) and have the purpose of guaranteeing maximum dose to the tumor, while maintaining minimum healthy tissue dosage so to limit undesirable side-effects. IGRT uses repeated 2D or 3D X-ray images from a patient in order to monitor a given RT treatment and, if necessary and possible, re-plan it (adaptive RT - ART) .

In a fractionated RT treatment course it is desirable to reposition the patient exactly in the same position as the one obtained in RT planning. However, small positioning errors can occur between sessions (interfractional mispositioning) and even during a session treatment ( intrafractional mispositioning) .

There are a number of degrees of freedom which may change between sessions. Consequently, finding an exact position becomes a complex challenge.

A common way to verify the patient positioning is recurring to two-dimensional (2D) imaging. Daily, in the beginning of each treatment weak, or when appropriate, megavoltage (MV) or kilovoltage (kV) radiographs of the patient may be acquired. Newer procedures include daily MV radiograph acquisitions in order to reduce positioning errors between sessions. This kind of imaging has poor image quality and detects only bony structures, i.e., soft tissue can be visualized poorly. Furthermore, 2D imaging is not fully able to detect rotational movements occurring, for example, in head-and-neck treatments (Nath et al . 2009). Such limitations in monitoring patient positioning can be overcome by volumetric or three-dimensional (3D) in-room imaging. CT on- rails is such a 3D imaging technique. It consists of a conventional CT scanner placed over a rail inside the treatment room. CT scanner and a linear accelerator (linac) share the same patient couch.

Although this system has the advantage of acquiring high-quality 3D images (capable of differentiating soft tissues), it requires a large room and more time to perform the procedure, and it is not able to detect intrafractional motions (Nath et al. 2009. Cone-beam CT (CBCT, which can be kV or MV) is also a 3D in-room imaging technique. CBCT consists mainly of a smaller CT scanner incorporated inside the treatment machine. Simultaneously with the treatment, images are acquired and position errors of the order of millimeters can be detected. An important issue of this method is the quality of the images obtained. Due to the scattering radiation in the target, the images obtained by CBCT have lower signal-tonoise ratio (SNR) than the images obtained by conventional CT (van Elmpt 2009) . Comparing the two CBCT modalities, kV CBCT has higher contrast and lower SNR than MV CBCT.

In helical tomotherapy (HT) the radiation is delivered in the tumor from many directions. This is a combination of a 6-MV. CT with a linac enclosed into a ring gantry (Tom'e et al. 2007). During the treatment, the radiation source in the gantry rotates around the patient, while the patient couchmoves along the inner hole of the gantry. HT allows acquiring daily volumetric images in order to visualize patient position errors and tumor/organ variations.

Digital tomosynthesis (DTS) is another 3D in-room imaging method. In this technique, the radiation source moves in an arc around the target (i.e., patient) and several 2D crosssection images are acquired. These cross-sectional images are used to reconstruct 3D images of the object being scanned.

As DTS constructs images from a limited number of arcs, it may result in lower deposited doses comparing to the other methodologies (Nath et al. 2009), although at the expense of inferior or incomplete datasets.

Volumetric imaging is superior to 2D imaging in many ways. But there is disagreement within the clinical community regarding to- 3D imaging. Recent studies, for example, state that the extra dosage due to either kV or MV CBCT is responsible for the emergency of secondary malignancies (Kan et al. 2008). Therefore, the frequency with which these exams must be performed is questioned. Some defend that weekly or biweekly scans are adequate, while others argue that daily scans are necessary.

Adaptive RT (ART) adjusts the dosage delivery plan to consider anatomical changes that can occur along the course of treatment (Schwartz and Dong 2011) . In this way, ART may guarantee maximum dose exposure to the tumor, maintaining minimum healthy-tissue dosage. A study of regression rate of tumors during treatment is presented by Barker Jr. et al. (2004) . They found that the gross tumor volumes (GTV) decreased at a median rate of 1.8% per treatment day, which corresponds to a median total relative loss of 69.5% in respect to the last treatment day. Moreover, they observe that the center of mass of shrinking tumors changed position with time, indicating that GTV loss was frequently asymmetric.

Volume of parotid glands also decreased at a median of 0.19 cm³ per day. This and other studies (Kuo et al. 2006; Hansen et al. 2006; 0' Daniel et al. 2007; Blanco et al. 2005; Vakilha et al. 2007) provide evidence that ART is an asset in the RT field. However, some questions need to be clarified, namely, if, when, how, and for which pathologies (Sterzing et al. 2011) the re-planning should be done, and with which frequency. In particular, studies have estimated that daily CBCT imaging can lead to an increase of dose of 5.3-6.7 cGy to skin per scan (Kan et al. 2008; Nath et al. 2009) and a total of 300 cGy over an entire treatment course (Ding et al. 2008; Nath et al. 2009) . This may correspond to a 2%-4% increase in secondary malignancies (Kan et al. 2008; Nath et al. 2009) . No long term data on the actual incidence of secondary malignancies is currently available, and continued investigation will have to be performed to address this question (Nath et al . 2009).

In summary, modern radiotherapy guidance techniques will profit, like in the CT diagnostic and screening scenario, from an imaging technology like OrthoCT, reducing dose with, again, no compromise at the level of image quality.

2 Concept and apparatus

2.1 Two-dimensional information from ID scanning

2.1.1 Concept: ID beam scan provides 2D information

The concept and principle of operation of orthogonal computed tomography (OrthoCT) is depicted in Fig. 1. The system shown is composed at least by a photon source capable of providing scanned photon beams, and one or more detector heads (two shown) each equipped with a collimator that may be multi-hole or multi-slit. The purpose is to detect radiation scattered (e.g. Compton) or emitted (e.g. pair production) from the target (e.g. patient) that forms right angles with the beam direction.

For each beam scan along the target in a certain direction 9_Ph₀ton source _? a scan in the opposite direction is performed:

Qphoton source ± n . The scan profile for a position is then obtained by the average of the profiles acquired in opposite directions .

2.1.2 Photon source rotation for enhancing image contrast

The scan along the phantom can also be acquired at different

Θ angles (with Θ ranging from 0 to 360 degrees), in an attempt to enhance the contrast of the obtained images if and when necessary. This may be done due to the fact that a given combination of complementary scans (Θ and θ+π) deposits a dose of the order of mGy or less, as shown in section 3. For this purpose, the setup photon source + detector heads depicted in Fig. 1 may rotate around the target, as sketched in Fig. 2.

2.1.3 Concomitant non-overlapping opposite beams

The setups depicted in Figs. 1 and 2 show opposite beams, as necessary for OrthoCT with double-sided irradiation. Although from the point of view of the detection system (s) such opposed beams may be turned on concomitantly, image quality as well as the operation and the durability of the photon sources may be affected if overlapping, opposed beams are turned on concomitantly. On the other hand, the operation of concomitant opposed beams allows for reducing twofold the scan time. For this purpose, a setup like the one depicted in Fig. 3 may be utilized. It allows for two non- overlapping scans in opposite directions to be performed by means of appropriate shielding. As shown, such shielding is mounted with dynamic beam ports at each photon source.

2.2 Three-dimensional information from 2D scanning

The OrthoCT concepts presented in section 2.1 regard the feasibility of obtaining 2D target information from ID beam scanning. In this section we extend that concept in order to obtain complete 3D information from the target, clinical, biological, or industrial. Due to the low dose impact of this technique, a clinical example will be presented.

2.2.1 Voxel-by-voxel beam 2D scanning

Fig. 4 depicts a beam being scanned through a 2D grid that trespasses the 3D object space (a human head in this example) .

This allows a pencil-beam-like irradiation which, consequently, benefits from the usage of a sliced detector collimatio .

2.2.2 Slice-by-slice beam scanning

The beam may also be delivered on a slice-by-slice manner, e.g. by irradiating voxels 0 to 9 in Fig. 4 concomitantly. Here, a multi-hole collimator positioned in front of the 2D detector system, and substituting therefore the sliced collimation system depicted, would be necessary. This is because of the need to identify the origin of each detected photon in respect to the slice being irradiated.

3 Simulation results

In this section we show results obtained by means of Geant4 (Agostinelli et al. 2003; Allison et al. 2006) . These results highlight the feasibility and high potential of the OrthoCT system being proposed. The simulation assumed, at this stage, a linac with endpoint energy of 6 MeV. This energy is commonly used in IMRT treatments of head-and-neck pathologies. Fig. 5 shows the simulated target phantom. It has cylindrical shape with 30-cm length and 18-cm diameter, and it is made mainly of water. Near the isocenter, a small air cavity is positioned, which represents any empty body region. In addition, a small cylinder made of cortical bone with 2-cm radius and 3-cm length is immersed in the middle of the phantom. The top image in Fig. 6 shows the profiles of photons that are collected at the annulus perfect detector, positioned at 10 cm from the isocenter, with a direction approximately orthogonal to the beam axis. For that, only photons arriving at the detector with an angle within the range 90±0.7 degrees (Fig. 5) and energy greater than 300 keV are counted.

This allows building each profile a and b. Profile a (b) regards the configuration where the beam enters from the left (right) . Profile c is the computed average of the a+b counts per millimeter. It is this profile (c) that is highly correlated with target density. Indeed, the number of collected counts per millimeter changes from 4000 in air, to 8000 in water, to 12000 in cortical bone. We highlight the fact that conventional CT scanners present their final results in Hounsfield units (HU) ranging between -1000 HU (air) to 0 HU (water) to 1000 HU (bone), a 2000 HU dynamic range (Pedroso de Lima 2009) . Some modern scanners have a greater range of HU up to 4000 (Jackson and Thomas 2004) .

In addition to this potential OrthoCT contrast enhancement, the bottom image in Fig. 6 shows that the average (maximum) dose corresponding to the radiation used to build profile c is 761 μ Gy (865 μ Gy) , which is less than a factor 40 in respect to the minimum dose utilized to perform a CT scan of the head of an adult (Huda and Vance 2007) .

It must be stated, in regards to the quality of the final profile (or image) obtained, that buildup effects that seem to lead to a non-flat count profile at the margins of the image, as observed in the counts profile c in the top image in Fig. 6, are highly suppressed when a real collimator is implemented, as shown experimentally by Sim0es et al . (2011) for a single-beam irradiation only.

In regards to target heterogeneity along the direction orthogonal to the beam, it must be equally stated that real imaging scenarios will not profit from the target azimuthal symmetry simulated in the setup of Fig. 5. This does not pose any limitation to OrthoCT. This is because the physicomathematical procedure intrinsic to the scan may start its computation from datasets that correspond to data collected from superficial target regions. Such regions are hence independent from any interior target heterogeneity since only air intercepts the superficial imaging region and the detector.

Three-dimensional image construction may then follow onto more interior target regions, where corrections for more superficial target heterogeneousness are already possible due to the previously-computed, superficial target densities . 4 Discussion and conclusions

A method and associated technology and instrumentation for obtaining 3D information from an irradiated target is presented.

First simulated results are presented, which use a beam energy much superior to that utilized in X-ray imaging and any CT technology- Despite this fact, positioning one or more collimated detectors orthogonally to the beam direction reveals that obtaining 3D target information is possible with a dose-reduction factor of 40 or more in respect to a CT of the brain of an adult. Such dose reduction is highly desired in the medical community operating in the fields of diagnostics, screening, or radiation therapy monitoring and/or guidance.

Scan timings associated with OrthoCT may provide an additional advantage since this technique does not need to rotate around the target/patient. Nevertheless, in situations where extreme image contrast is necessary, such rotation of the whole setup source (s) + detector (s) may provide valuable high-quality imaging and is, therefore, not discarded.

Furthermore, imaging rotation timings in the radiotherapy environment is limited by legislated regulations to one or less rpm (rotation per minute) due to the bulky and heavy instrumentation lying nearby the patient. This represents indeed an increased added value to OrthoCT in such environment due to its rotation-free potential: faster scanning, highly reduced dose, and increased patient throughput at a decreased potential for induction of secondary malignancies.

We finally state that the dose associated with the OrthoCT technology being proposed here lies below the mGy level, which renders it competitive even with X-ray imaging.

Therefore, an additional huge market may be present when such technology is brought into the clinical X-ray screening environment of both mammography and radiographic procedures, with typical mammography dose lying between 1.1 to 2.0 mGy (Hufton 2002) and, for example, the entrance surface dose in an anterior/posterior or posterior/anterior skull radiographic procedure being 3 mGy (Hufton 2002) .

Figure captions :

Fig. 1. Principle of (OrthoCT) (1, 4). A beam (7) coming from at least one photon source (10) irradiates the target (11) (e.g. a patient). Radiation scattered or re-emitted from the target (11) and escaping with a right angle in respect to the beam direction is collimated (2, 5) and collected by a detection system (3, 6) (Head 0 and/or Head 1). For each angle Q hoton source corresponding to the position of the photon source, a complementary scan with the photon source in the opposite direction (9_Ph₀ton source+n) may be performed.

Fig. 3. Layout of the photon source (10) and detector head(s) (1, 4) rotating around the target (11). This acquisition at different 9_Ph₀ton source angles can enhance the contrast of the obtained images.

Fig. 2. Setup for simultaneous, non-overlapping scans in opposite directions. The beam position of each photon source (10, 13) is controlled by appropriate active shielding (12, 14), ensuring the safe operation of two concomitant beams (7) in opposite directions, not disturbing one another. This configuration leads to a scanning system potentially two times faster than those depicted in Figs. 1 and 3.

Fig. 4. OrthoCT in a clinical scenario. The beam (7) is scanned through a 2D grid (9) that trespasses the 3D object space (a human head (8) in this example). This allows a pencil-beam-like irradiation which, consequently, benefits from the usage of a sliced detector collimation (2, 5) . Should the beam (7) be delivered on a slice-by-slice manner, e.g. by irradiating voxels i to x concomitantly, than a multi-hole collimator would be necessary (section 2.2.2).

Fig. 5. Setup simulated with Geant4 for first OrthoCT evaluation .

Fig. 6. Geant4 simulation results obtained with the setup of Fig. 5. Top: Count profiles obtained with the beam impinging on the phantom from the left (dashed line, a) and right (solid line, b) ; and average counts (dotted line, c) . Bottom: Dose profiles corresponding to each irradiation scenario presented in the top image.

LEGEND:

1. OrthoCT device

2. Sliced collimator

3. X- and gamma ray detector

4. Second orthoCT device (optional)

5. Sliced collimator of second orthoCT device

6. X- and gamma ray detector of second orthoCT device

7. Thin X-ray beam

8. Patient (target being imaged)

9. Pixel grid computed by software for patient beam scanning

10. X-ray source (LINAC or other)

11. Target

12. Shielding with dynamic beam port selection (example multileaf collimator)

13. Second X-ray source (optional)

14. Second shielding with dynamic beam port selection

15. Air cavity

16. Cortical bone 17. Annulus perfect detector

a. Simulation results (irradiation from left)

b. Simulation results (irradiation from right)

c. Simulation results (left and right irradiation)

i, ii, iii, iv, xv. 2D pixel grid superposed over imaging target (patient)

References

S. Agostinelli, J. Allison, K. Amako, J. Apostolakis, H. Araujo, et al . Geant4 : A simulation toolkit. Nucl . Instrum. Meth. A , 506 (3) :250-303, 2003.

J. Allison, K. Amako, J. Apostolakis, H. Araujo, P. A. Dubois, et al. Geant4 developments and applications. IEEE Trans. Nucl. Sci. , 53(l):270-8, 2006.

A. Alves, C. Rebelo, and .C. Lopes. Radioterapia no IPO de Coimbra, 2008. Informational flyer from the Instituto Portugues de Oncologia de Coimbra Francisco Gentil, E.P.E , in Portuguese.

I. Andersson and L. Janzon. Reduced breast cancer mortality in women under age 50: Updated results from the Malm^'O mammographic screening program. Monogr. Natl. Cancer Inst., 22:63-7, 1997. J.L. Barker Jr., A.S. Garden, K.K. Ang, et al. Quantification of volumetric and geometric changes occurring during fractionated radiotherapy for head-and-neck cancer using an integrated CT/linear accelerator system. Int. J. Radiat. Oncol. Biol. Phys., 59 (4 ): 960-70, 2004.

A.I. Blanco, K.S.C. Chao, I. El Naqa, et al . Dose-volume modeling of salivary function in patients with head-andneck cancer receiving radiotherapy. Int. J. Radiat. Oncol. Biol. Phys. , 62 (4) : 1055-69, 2005.

D. J. Brenner and E.J. Hall . Computed tomography - An increasing source of radiation exposure. N. Engl. J. Med., 357:2277-84, 2007.

G.X. Ding, D.M. Duggan, and C.W. Coffey. Accurate patient dosimetry of kilovoltage cone-beam CT in radiation therapy. Med. Phys , 35 ( 3 ): 1135-4 , 2008.

E. K. Hansen, M.K. Bucci, J.M. Quivey, V. Weinberg, and P. Xia. Repeated CT imaging and replanning during the course of IMRT for head-and-neck cancer. Int. J. Radiat. Oncol. Biol. Phys. , 64 (2) :355-62, 2006.

W. Huda and A. Vance. Patient radiation doses from adult and pediatric CT. A.J.R. , 188:540-6, 2007. A. P. Hufton. Practical Radiation Protection in Health Care, chapter 14: Diagnostic radiology: patient dosimetry, pages

221-41. Oxford University Press, New York, 2002. ISBN 0-19- 263082-2. (C.J. Martin and D.G. Sutton, Eds.).

S.A. Jackson and R. . Thomas. Cross-Sectional Imaging Made Easy, chapter 1: Introduction to CT physics, pages 3-16. Churchill Livingstone, London, 2004. ISBN 044307187X.

M.W.K. Kan, L.H.T. Leung, W. Wong, and N. Lam. Radiation dose from cone beam computed tomography for image-guided radiation therapy. Int. J. Radiat . Oncol. Biol. Phys . , 70(l):272-9, 2008.

Y.-C. Kuo, T.-H.Wu, T.-S. Chung, et al. Effect of regression of enlarged neck lymph nodes on radiation doses received by parotid glands during intensity-modulated radiotherapy for head and neck cancer. Am. J. Clin. Oncol. , 29(6): 600-5,

2006.

S.K. Nath, D.R. Simpson, B.S. Rose, and A. P. Sandhu. Recent advances in image-guided radiotherapy for head and neck carcinoma. J. Oncol. , 2009. doi : 10.1155/2009/752135.

J.C. O'Daniel, A.S. Garden, D.L. Schwartz, et al . Parotid gland dose in intensity-modulated radiotherapy for head and neck cancer: is what you plan what you get? Int. J. Radiat. Oncol. Biol. Phys . , 6 ( 4 ): 1290-6, 2007.

J.J. Pedroso de Lima. T' tecnicas de Diagnostico com Raios X - Aspectos F'isicos e Biof'isicos , pages 5-7. Coimbra University Press, second edition, 2009. ISBN 978-989-8074-83- 6. In Portuguese.

D.L. Schwartz and L. Dong. Adaptive radiation therapy for head and neck cancer - can an old goal evolve into a new standard? J. Oncol. , 2011. doi: 10.1155/2011/690595.

H. Simoes, M. Pinto, M. Cunha, J. Gon.calves, L. Sampaio, R.J. Ferreira, H.M. Saraiva, A.R. Barbeiro, M. Capela, B. Ferreira, P. Fonte, S. Ghithan, A. Leal, M.C. Lopes, P. Martins, and P. Crespo. Dose-free monitoring of radiotherapy treatments with scattered photons: First experimental results at a 6-MV linac. In Conf. Records 2011 IEEE Nucl. Sci. Symp . & Med. Imag. Conf. , Valencia, Spain, 23- 29 Oct. 2011. MIC21.S-258.

F. Sterzing, R. Engenhart-Cabillic, M. Flentje, and J. Debus. Options of image-guided radiotherapy - a new dimension in radiation oncology. Dtsch. Arztebl. Int., 108 (16) : 274-80, 2011.

W.A. Tome, H.A. Jaradat, I. A. Nelson, M.A. Ritter, and M.P. Mehta . Helical tomotherapy: image guidance and adaptive dose guidance. Front. Radiat. Ther. Oncol. , 40:162-78, 2007.

M. Vakilha, D. Hwang, S.L. Breen, et al. Changes in position and size of parotid glands assessed with daily conebeam CT during image-guided IMRT for head and neck cancer: implications for dose received. Int. J. Radiat. Oncol. Biol. Phys., 69:S438-S439, 2007.

W. van Elmpt. 3D dose verification for advanced radiotherapy . PhD thesis, Department of Radiation Oncology, School for Oncology and Developmental Biology, Maastricht University Medical Center, Maastricht, The Netherlands, March 2009.

Claims

1. Orthogonal computed tomography system (1) for clinical diagnostics and screening, image guided radiotherapy and adaptive radiotherapy, and other tomographic applications in medicine and industry, which comprises at least one photon source (10, 13) capable of providing scanned photon beams and one or more detector heads (1, 4) each equipped with a collimator (2, 5) that may be multi-hole or multi-slit, wherein the axis of the photon source and that of the detector head or heads form an angle of 90 degrees.

2. Orthogonal computed tomography system (1) according to claim 1 wherein the detector is an X-and/or a gamma ray detector.

3. Orthogonal computed tomography system (1) according to the previous claims wherein, for each beam scan along the target (8, 11) in a certain direction e_photon source _/ a scan in the opposite direction may be performed.

4. Orthogonal computed tomography system (1) according to the previous claims wherein the scan profile for a position is obtained by irradiating the target from one direction only.

5. Orthogonal computed tomography system (1) according to the previous claims wherein the scan profile for a position is obtained by combining profiles acquired in

Opposite directions (Θ photon source and Θ photon source ±n) .

6. Orthogonal computed tomography system (1) according to the previous claims wherein the photon source (10) or sources (10, 13) and the detector head (1) or heads (1, 4) may rotate around the target.

7. Orthogonal computed tomography system (1) according to the previous claims wherein the non-overlapping scans in opposite directions is performed by means of a shielding (12, 14) .