|
|
||||||||
|
|
|||||||||
|
|
Instrumentation & Data Analysis: Image GenerationPET - Motion Effects and Compensation |
1 PET Imaging Centre, Guy's, King's, and St. Thomas' School of Medicine, London, United Kingdom
238
Objectives: Respiratory gating is a technique used to acquire an image as a series of discrete frames which are free of respiratory motion. Typically additional hardware is required which monitors the respiratory state and generates a trigger signal. The aim of this work was to develop a retrospective, automated PET and CT respiratory gating technique which is data driven, such that it derives a respiratory signal and trigger sequence from the data itself.
Methods: Low dose (10 mAs) CT images were acquired in cine mode, and 5 min PET images acquired in list mode over a single 16 cm position. A signal correlated with respiratory state was derived directly from the PET and CT datasets by a multi-step procedure based on spectral analysis. The signal was used to bin CT and list mode PET data into 12 independent frames. Sinograms were attenuation corrected using matching CT frames, and reconstructed. Lung images of 4 patients with lung cancer were acquired to evaluate the technique. A hardware based respiratory monitoring system was also used to record the respiratory state of the patients during both acquisitions.
Results: A signal reflecting respiratory state was successfully extracted from cine CT and list mode PET data, and used to gate the acquisition retrospectively. Respiratory motion of the lung and lung nodules was clearly visible in all PET and CT image sequences. The data derived respiratory signals corresponded well with the hardware acquired signals.
Conclusions: An automated, retrospective technique for respiratory gating PET and CT acquisitions has been developed. The system was found to operate successfully on PET/CT patient images of the lungs using cine CT acquisitions and list mode PET acquisitions.
| ||||||||||||||||||||||||||||||||||||||