Current status of diagnostic counting and imaging techniques used in nuclear medicine: A sketch

Diagnostic procedures used in Nuclear Medicine are designed to provide information about the static and/or dynamic distribution of some particular stable or radioactive material within the patient, as well as the quantity of the material present. Thus, the static distribution of ¹³¹I in the thyroid indicates the structure of the gland, whereas changes in the distribution or in the quantity present indicate its rate of functioning. A remarkable variety of techniques, based on emission, transmission, fluorescence, and incoherent holographic imaging principles, have been developed. Both moving and fixed radiation detectors are used, and these produce planar and 3-D projections and transverse sections of the object of interest. In general, the image quality is determined by detector characteristics such as sensitivity, spatial resolution, and depth of field, as well as by the examination time, object contrast, and object intensity, which is always severely limited in the interest of a low radiation dose to the patient. Image processing techniques can be used to improve image quality by removing certain artifacts and by producing a degree of sharpening that is limited by the signal-to-noise ratio. Incoherent holography and optimum-weighted multichannel imaging are presented as two approaches to increasing the signal-to-noise ratio. Alternative criteria for the evaluation of imaging procedures are presented, with an emphasis on the use of ROC curves to describe observer performance.