Lumbar myelography (A) Oblique view of a normal myelogram. The cerebrospinal fluid (CSF) is enhanced following intrathecal injection of water-soluble, radio-opaque contrast material. The nerve roots are seen as elongated thread-like filling defects running through the enhanced CSF and emanating from the dural sac. Note the roots pushed away by the needle tip. (B) Drop metastasis are demonstrated as rounded filling defects of various sizes, scattered along the cauda equina in this pathological myelogram. (C) Myelo-CT at the atlantoaxial level showing the enhanced CSF in the subarachnoid space. Note the dark—non-enhanced—oval shape spinal cord in the center and bilateral horizontally oriented delicate roots. (D) Sagittal reconstruction of myelo-CT at the C1-C2 level. (E) Coronal reconstruction of myelo-CT through the whole cervical spine. (F) Axial cut through the tip of the conus medullaris at L1 level.
for scanning. Scans are made at steps of 10°, which corresponds to the angle of the fan beam. The minimum scan time ranges from 6 to 20 seconds.
3. Rotation scanner with movable detectors (third generation). A broad fan beam penetrates the test object and rotates around the body along with a detector array containing 200 to 1,000 detector units. The minimum scan time is 1 to 4 seconds.
4. Rotation scanner with stationary detectors (fourth generation). The angle of the fan X-ray beam covers the entire test object. The X-ray source rotates inside or outside a stationary ring detector array with 300 to 4,000 detectors in order to scan the test object. The scan time ranges from 3 to 8 seconds.
Short scanning times are desirable in spinal computed tomography, because motion artifacts can thereby be eliminated. Slow scanning systems with alternating, contrarotating movements are therefore being replaced by continuously rotating systems with faster scanning times. The attenuation values for each set of projections are registered in the computer, and the CT image is reconstructed by means of a complex computational process. The finite number of attenuation values corresponding to the scanned object is organized in matrix form. The translation of these numbers into various analogous gray levels creates a visual image of the scanned cross-sectional area. Due to their different absorptive capacities, different internal structures will be identifiable on the picture image. The size of the image matrix, more specifically the number of calculated picture elements, is dependent on the number of individual projections. Matrix size, therefore, also influences the quality of image resolution.
The smallest unit of a computed tomogram is the individual picture element, or pixel. A pixel represents a certain proportion of the total cross-sectional area or a tissue element the volume of which is determined by the slice thickness, matrix size, and diameter of the scanning field. Under these conditions, a picture element also represents a volume element, or voxel.
Each volume element is given a numerical value, an attenuation value, which corresponds with the average amount of radiation absorbed by the tissue in that picture element. CT density is directly linearly proportional to the attenuation coefficient, a tissue-dependent constant influenced by many factors. The attenuation coefficient quantifies the absorption of X-irradiation. In CT, attenuation values are measured in Hounsfield units (HU). The attenuation value of air and water (defined as -1,000 HU and 0 HU, respectively) represent fixed points on the CT density scale that remain unaffected by tube voltage. Depending on the effective radiation of the scanning device, the relationship of the attenuation of different tissue types to water attenuation will vary. Density values listed in the literature (Table 2-1) must therefore be considered as mere guidelines.
Was this article helpful?