A phantom is an artificial object of known size and composition that is imaged to test, adjust or monitor the linearity, sensitivity, homogeneity, imaging performance and geometric aspects of CT scanners, gamma cameras or other radiographic systems. A phantom can be a fluid-filled container or bottle including structures of various materials, sizes and shapes.

See also Daily Quality Assurance and Acceptance Checking.

A calibration is a correction procedure that determines the relationship between the measured output of a system and the reference standard. Calibration procedures include scanning air or an appropriate test phantom.
The calibration of a CT system takes account of variations in beam intensity or detector efficiency in order to achieve best homogeneity within the field of view and the accuracy of CT numbers.

See also Calibration Factor and Acceptance Checking.

Grids are used in a test phantom to check the quality of x-ray images. Such a grid may consist for example, of gold structures on a silicon surface and a grid period length of 2 micrometer with space wide in between of 1 micrometer.
Filter grids are also used in x-ray imaging to reduce noise contributed by scatter.

Noise is an undesirable background interference or disturbance that affects image quality. The image noise is commonly characterized by the standard deviation of CT numbers in the image of a uniform object (phantom - generally water) relative to the difference in CT numbers between water and air.
Noise is an important limiting factor of CT image quality and is characterized by a grainy appearance, sometimes described as a salt and pepper pattern. Noise as a result of low radiation dose requires a higher tube current (mA) in combination with narrow slices to maintain image noise at an acceptable level.

(PSF) The point spread function describes the response of an imaging system to a point or impulse input. A hypothetical point object will generally have an extended (blurred) image resulting from the imaging process; this is the point spread function characterizing the imaging process. Considering any object as composed of an assembly of point objects, knowledge of the PSF permits the prediction of how the object will be imaged, assuming linearity of the imaging process.

See also Resolution, Phantom and Daily Quality Assurance.