Computed Tomography, Patient-Based Artifacts - Essay Example

Introduction Computed Tomography (CT) is extensively used in the field of imaging to study the geometry, composition and mass density of interior organs by taking two-dimensional x-ray projections of the organs from which 3D images are reconstructed (Yazdi & Beaulieu, 2008; Barrett & Keat, 2004). Presence of an artifact compromises the quality of the image generated from the scan, thus making it difficult to use the image as a diagnostic tool (Barrett & Keat, 2004). There are several ways by which artifacts originate which include processes involving CT data acquisition, movement of patients during the scanning process or presence of metals on the body, problems with the scanner or during the reconstruction process (Barrett & Keat, 2004). The various types of artifacts and the methods used for reducing their occurrence are discussed below. Physics-based artifacts These are mainly concerned with acquisition of CT data. Problems with the detector or other parts of the scanner can give rise to artifacts in the final image. Beam Hardening The most common physics-based artifact is beam hardening which is caused due to differential absorption of low and high energy photons of an x-ray beam as it passes through an object. Such beam hardening phenomena can give rise to cupping and streaking or dark band artifacts. Cupping artifacts arise when the x-rays passing through the middle portion of an object become hardened which causes a reduction in its attenuation rate resulting in an intense beam reaching the detector. Streaking is another common phenomenon where streaks or dark bands appear between two dense objects again due to the hardening effect. This is usually visualized in scans taken in bony regions (Barrett & Keat, 2004). The methods used to reduce artifacts due to beam hardening include filtration of low energy particles, calibration correction, and use of appropriate software algorithms to correct the hardening (Barrett & Keat, 2004; Huang, n.d; Petit et al, 2010). Partial volume Presence of a heterogeneous tissue mix can result in a CT number that is an attenuation average of all tissue types which can in turn result in a partial volume artifact as bands or streaks. Presence of off-axis objects in the path of the x-ray beam can result in the appearance of shading artifacts in the scan image. Such artifacts can be avoided using thinner sections and image noise can be limited by combining thinner sections to form a thicker section (Barrett & Keat, 2004; Huang, n.d). Photon starvation This effect occurs in parts of the body where attenuation of the x-ray beam is greatest such as the shoulders and the hip. This results in low number of photons reaching the detector which causes noisy projections that are in turn magnified during the reconstruction process. Streaking artifacts in the image result due to such high magnification (Yazdi & Beaulieu, 2008; Barrett & Keat, 2004). This effect can be corrected by increasing the tube current. However this would cause the patient to be exposed to an extra dosage when the beam passes through less attenuating regions. Newer technologies have now helped to alleviate this unwanted exposure. Undersampling The number of projections and the interval between them is vital for reconstructing a CT image. Large intervals between projections can result in misregistration artifacts wherein fine stripes radiate from the borders of dense structures, an effect referred to as view aliasing. In some cases the stripes may appear to originate from within a projection and are found close to the structure, an effect referred to as ray aliasing. Though the appearance of anatomic structures are not affected by the presence of stripes, in cases where accurate detailing is required such artifacts should be avoided. This is generally done by obtaining an increased number of projections in a single rotation. Ray aliasing effects can be reduced by using more high-resolution techniques (Barrett & Keat, 2004). Patient-based artifacts Metal artifacts Metallic objects present in the scan field can cause increased attenuation of the x-ray beams, due to their high density, resulting in streaking and beam hardening artifacts (Yazdi & Beaulieu, 2008; Barrett & Keat, 2004; Huang, n.d). Examples of metallic implants include hip prostheses, dental fillings and clips (Yazdi & Beaulieu, 2008; Huang, n.d). While generally metallic objects are removed prior to a CT scan presence non- removable metallic implants such as dental fillings are avoided in the scan by employing methods such as gantry angulation (Barrett & Keat, 2004; Lewis et al, 2010). In a study carried out by Lewis et al, the metal artifacts resulting from total knee replacements were significantly reduced by gantry angulation process. In this the CT data was acquired by gantry angulation at 0, 5, 10 and 15 degree directions. The results showed a displacement of the artifacts from the areas of interest rather than reduction (Lewis et al, 2010). In case of surgical clips, a study carried out by Silverman et al showed that use of titanium clips effectively reduced the artifacts produced compared to tantalum and stainless steel clips and when coupled with a faster scan time the amount of clip artifacts produced were reduced (Silverman et al, 1986). When the area involving the metallic object is to be included in the scan more high-end techniques such as kilovoltage that would penetrate the object can be used (Barrett & Keat, 2004). Other methods for metal artifact reduction employed include software’s that are based on the projection matrix (Yazdi & Beaulieu, 2008; Barrett & Keat, 2004; Huang, n.d). The use of metal applicators during intracavitary brachytherapy for the treatment of cervical cancer gives rise to metal artifacts during CT scanning. While the projection-interpolation data has been widely used for hip prostheses and surgical clips, the study carried out by Roeske et al successfully used the algorithm to reduce metal artifacts in CT images taken after brachytherapy (Roeske et al, 2003). Patient movement Movement of patients during the scanning process gives rise to misregistration artifacts that appear as streaks in the reconstructed image. While voluntary motion of patients can be controlled by using aids, involuntary movements need to be managed by reducing the scanning time, or by use of sedation. In case of respiration the patients is advised to hold their breath for the duration of the scan. Additionally scanners also possess in-built features to minimize motion artifacts such as extra scanning modes or under-scanning modes that add or reduce rotation either at the beginning or at the end of a scan where maximum artifacts appear. In order to minimize artifacts produced due to rapid heart motion, which can give rise to artifacts, techniques referred to as gating by which data is collected from parts of the cardiac cycle where there is least cardiac motion is employed (Yazdi & Beaulieu, 2008; Barrett & Keat, 2004; Huang, n.d). Correction algorithms for the reconstruction process have also been proposed for motion artifact removal (Yazdi & Beaulieu, 2008; Huang, n.d). Motion artifacts of brain images have also been reduced by the process of segmental reconstruction of partial images from the raw data (Lavdas et al, 2009). More recently reformatting software that views images in various plane apart from the plane of acquisition has been developed which has helped to reduce movement related artifacts (Crocker et al, 2009). Scanner based artifacts Circular and helical artifacts Circular or ring artifacts at each angular position will appear when detectors are not properly calibrated. The presence of such ring artifacts would not damage the anatomical details of a scan image; however it will impair the diagnostic quality of the image as problems with the detector will create a dark smudges at the center of the image. This is generally corrected by repairing the detector system and recalibration which is now carried out by using suitable software’s (Barrett & Keat, 2004). The changing nature of anatomic structures in the z direction can give rise to helical artifacts due to helical interpolation and the artifact orientation varies depending on the position of the tube in the center of the plane. These artifacts can be reduced by using thin section of the material and by choosing a low pitch helical interpolator. Z-filter interpolators are most commonly used in multi-section scanners which help to reduce artifacts (Barrett & Keat, 2004). Conclusion In conclusion, there are various ways by which artifacts originate and their presence compromises on image quality and reduces the diagnostic use of the scan image. Thus it is important to identify these artifacts and reduce them completely or to the bare minimum in order to facilitate better use of the scan data for treatment purposes. Reference 1. Yazdi, M. & Beaulieu, L. (2008). Artifacts in spiral X-ray CT Scanners: Problems and Solutions. International Journal of Biological and Life Sciences. [Online] 4 (3), 135-139. Available from http://www.waset.org/journals/ijbls/v4/v4-3-24.pdf [Accessed 11th August 2010] 2. Barrett, J.F & Keat, N. (2004). Artifacts in CT: Recognition and Avoidance. Radio Graphics. [Online] 24 (6), 1679-1691. Available from http://www.imre.ucl.ac.be/rpr/RDGN3120/CT_artifact_Radiographics_Barrett.pdf [Accessed 11th August 2010] 3. Lavdas, E., Vlychou, M., Roka, V., Wozniak, G., Protogerou, G. & Fezoulidis, I. (2009). Artifacts in spiral CT protocols: The importance of the spatial reconstruction. European Journal of Radiography. [Online] 1, 73-79. Available from: doi:10.1016/j.ejradi.2009.09.004 [Accessed 11th August 2010] 4. Huang, Y. (n.d). Artifacts in Helical CT images. [Online] Available from http://ric.uthscsa.edu/personalpages/lancaster/DI2_Projects_2003/HelicalCT_artifacts.pdf [Accessed 11th August 2010] 5. Petit, S.F., van Elmpt, W.J.C., Lambin, P. & Dekker, A.L.A.J. (2010). Dose recalculation in megavoltage cone-beam CT for treatment evaluation: Removal of cupping and truncation artifacts in scans of the thorax and abdomen. Radiotherapy and Oncology. [Online] 94, 359-366. Available from: doi:10.1016/j.radonc.2009.12.001 [Accessed 11th August 2010] 6. Roeske, J.C., Lund, C., Pelizzari, C.A., Pan, X. & Mundt, A.J. (2003). Reduction of computed tomography metal artifacts due to the Fletcher-Suit applicator in gynecology patients receiving intracavitary brachytherapy. Brachytherapy. [Online] 2, 207-214. Available from: doi:10.1016/j.brachy.2003.08.001 [Accessed 11th August 2010] 7. Lewis, M., Toms, A.P., Reid, K. & Bugg, W. (2010). CT metal artifact reduction of total knee prostheses using angled gantry multiplanar reformation. The knee. [Online] 17, 279-282. Available from: doi:10.1016/j.knee.2010.02.007 [Accessed 11th August 2010] 8. Silverman, P.M., Spicer, L.D., McKinney, R. & Feldman, D.B. (1986). Computed tomographic evaluation of surgical clip artifact: Tissue phantom and experimental animal assessment. Computerized Radiology. [Online] 10 (1), 37-40. [Accessed 11th August 2010] 9. Crocker, M., Matthews, H., Willis-Owen, C., Willis-Owen, S., Rich, P. & Minhas, P. (2009). Movement artifacts mimicking type 2 odontoid fracture on CT reconstructions. Injury Extra. [Online] 40, 63-64. Available from: doi:10.1016/j.injury.2008.11.007 [Accessed 11th August 2010] Read More