Encrypted login | home

Program Information

Performance Evaluation of Digital PET/CT: Medical Physics Basis for the Clinical Applications

no image available
J Zhang

J Zhang1*, M Miller2 , MV Knopp1 , (1) The Ohio State University, Columbus, OH, (2) Philips Healthcare, Highland Heights, Ohio

Presentations

SU-F-I-55 (Sunday, July 31, 2016) 3:00 PM - 6:00 PM Room: Exhibit Hall


Purpose: Replacement of conventional PMT-based detector with next generation digital photon counting (DPC) detector is a technology leap for PET imaging. This study evaluated the performance and characteristics of the DPC system and its stability within a 1 year time window following its installation focusing on the medical physics basis for clinical applications.

Methods: A digital PET/CT scanner using 1:1 coupling of 23,040 crystal:detector elements was introduced and became operational at OSU. We tested and evaluated system performance and characteristics using NEMA NU2-2012. System stabilities in timing resolution, energy resolution, detector temperature and humidity (T&H) were monitored over 1-yr. Timing, energy and spatial resolution were characterized across clinically relevant count rate range. CQIE uniformity PET and NEMA IEC-Body PET with hot spheres varying with sizes and contrasts were performed. PET reconstructed in standard(4mm), High(2mm) and Ultra-High(1mm) definitions were evaluated.

Results: NEMA results showed PET spatial resolution (mm-FWHM) from 4.01&4.14 at 1cm to 5.82&6.17 at 20cm in transverse & axial. 322±3ps timing and 11.0% energy resolution were measured. 5.7kcps/MBq system sensitivity with 24kcps/MBq effective sensitivity was obtained. The peak-NECR was ~171kcps with the effective peak-NECR >650kcps@50kBq/mL. Scatter fraction was ~30%, and the maximum trues was >900kcps. NEMA IQ demonstrated hot sphere contrast ranging from ~62%±2%(10mm) to ~88%±2%(22mm), cold sphere contrast of ~86%±2%(28mm) and ~89%±3%(37mm) and excellent uniformity. Monitoring 1-yr stability, it revealed ~1% change in timing, ±0.4% change in energy resolution, and <10% variations in T&H. CQIE PET gave <3% SUV variances in axial. 60%-100% recovery coefficients across sphere sizes and contrast levels were achieved.

Conclusion: Characteristics and stability of the next generation DPC PET detector system over an 1-yr time window was excellent and better than prior experiences. It demonstrated improved and robust system characteristics and performance in spatial resolution, sensitivity, timing and energy resolution, count rate and image quality.

Funding Support, Disclosures, and Conflict of Interest: Michael Miller is an employee of Philips Healthcare


Contact Email: