Open-Geometry PET: Quantifying the Trade-off Between Time-of-Flight Resolution and Angular Coverage

OBJECTIVE: Time-of-flight (TOF) resolution improves positron emission tomography (PET) by localizing annihilation events along the line of response, which reduces statistical noise and improves image quality. In open-geometry PET systems, improved TOF can compensate for limited angular sampling by reducing geometric distortions and reconstruction artifacts. This work quantitatively studies how TOF resolution and angular coverage together determine image quality in open-geometry PET. 
Approach: Using Monte Carlo simulations, we estimate the minimum angular coverage required to obtain reconstructions with minimal distortion for different TOF performance levels. An idealized PET scanner model is first used to isolate the effects of angular sampling and TOF, without detector-related blurring. Five angular coverages (60°, 90°, 120°, 150°, and 180°) and six coincidence timing resolutions (400, 200, 100, 75, 50, and 25 ps FWHM) are evaluated. Image quality is assessed using point sources, the Derenzo phantom, and the NEMA image quality phantom, reporting spatial resolution, contrast recovery, background variability, and geometric distortion quantified by ellipse fitting. 
Main results: The results show that TOF resolutions below approximately 100 ps are a key enabling factor for highly open PET geometries, substantially reducing limited-angle artifacts and allowing image quality to approach that of full-ring systems. The main trends are further validated using a realistic Monte Carlo model of the Siemens Biograph Vision PET/CT scanner. 
Significance: These results provide practical design guidance for next-generation open-geometry PET systems, showing that fast TOF performance can reduce required detector coverage while preserving image quality.