Optimal number of respiratory gates in positron emission tomography: A cardiac patient study

Respiratory gating is the method of dividing the data from a tomographic scan with respect to the respiratory phase of the patient. It enables more accurate images by reducing the effects of motion blur and attenuation artifacts due to motion. However, it induces image degradation due to higher noise levels as the number of events per gate is reduced. Due to lack of systematic studies in this regard, different numbers of gates are being used in the scientific and clinical practice. The present study aims at examining the relationship between the respiratory signal, the number of gates required for accurate motion detection, and the level of noise with two different methods of gating: (1) Amplitude-based gating and (2) time-based gating. Patient data with a wide range of motion are used for the study. The results show that time-based gating underestimates the real respiratory displacement by up to 50%. The optimal number of gates is 8 for amplitude- and 6 for time-based gatings. The noise properties remain the same with either method but noise increases with increasing number of gates.