PT  - JOURNAL ARTICLE
AU  - Bin Zhang
AU  - Chi-hua Tung
TI  - Investigation of TOF 3D local tomography with various time resolutions using clinical datasets
DP  - 2010 May 01
TA  - Journal of Nuclear Medicine
PG  - 1382--1382
VI  - 51
IP  - supplement 2
4099  - http://jnm.snmjournals.org/content/51/supplement_2/1382.short
4100  - http://jnm.snmjournals.org/content/51/supplement_2/1382.full
SO  - J Nucl Med2010 May 01; 51
AB  - 1382 Objectives TOF provides local information which improves the noise behavior and spatial resolution of PET reconstruction. The goal of this paper is to investigate the clinical benefits from TOF 3D local tomography (i.e. ROI) with various time resolutions using clinic datasets. Methods Local tomography can benefit clinical PET imaging in many ways such as dealing with truncated datasets and/or reducing reconstruction time to improve workflow. In this paper, we investigate the TOF 3D local tomography properties for PET imaging through four comparison studies: 1) comparing image quality and quantitative accuracy between TOF-PET and nonTOF-PET ROI reconstructions with clinical datasets; 2) comparing computational cost between TOF ROI reconstruction and full TOF FOV reconstruction; 3) comparing TOF-PET and nonTOF-PET ROI reconstructions with truncated datasets; 4) comparing TOF-PET ROI reconstruction with different time resolutions. A list-mode ROI-based OSEM algorithm is developed and implemented based on a Philips Gemini TF system. This algorithm is used to reconstruct both the TOF and nonTOF clinical datasets. Mulitple truncated datasets are generated by limiting the scan angles and reducing FOV size. Two TOF time resolutions are investigated: 580ps (current Gemini TF) and 495ps (newly announced by Philips). Results One comparison result is shown in Fig. 1. TOF ROI images have better image quality than the nonTOF ROI ones. The bias between the TOF ROI image and the full-FOV image is much smaller. For the cardiac ROI study in Fig. 1, TOF ROI recon is more than 27% faster than nonTOF ROI, and 45% faster than the full FOV recon. TOF ROI recon achieves slightly better results with the 495ps time resolution. Conclusions Clinical 3D PET data can be redundant or insufficient. TOF ROI reconstruction solves this problem and provides useful ROI images with more efficient reconstruction. With improved time resolution, TOF-based ROI reconstruction may benefit PET imaging through many applications in the future such as patient local motion compensation, zoomed ROI reconstruction, or small FOV imaging