Improvement of Spatial Resolution With Iterative PET Reconstruction Using Ultrafast TOF

Abstract

The impact of time of flight (TOF) on positron emission tomography (PET) spatial resolution is generally considered negligible. In this work, a two-step approach based on simulations of 2-D scanner configurations is taken to show that ultrafast TOF has the potential to overcome the limitation induced by the physical size of detectors on spatial resolution. An estimation of the lower bound on spatial resolution using point sources is provided, followed by a qualitative assessment of the resolution obtained using a Hot Spot phantom. The impact of detector width, TOF resolution, and TOF binning on the achieved spatial resolution is also studied. While gain beyond the expected blur due to detector size is demonstrated, the detector size remains one limiting factor albeit less prominent. The dependence on acquisition statistics to reach the full potential of TOF-induced gain in spatial resolution is demonstrated. A simulated brain phantom acquired with a fictive 3-D PET scanner was qualitatively analyzed and structures smaller than the typical limit are clearly made visible by reconstructing the images with a ∼13-ps TOF resolution. A potential application of this feature of ultrafast TOF would be the design of clinical PET scanners achieving spatial resolution beyond the current state of the art.

Parameterizing the Angular Distribution of Emission: A model for TOF-PET low counts reconstruction

“…For both the PADE v0 model and the MLEM model, the 3.2 mm spots were mostly resolved even with detectors being 8 mm in width. This was expected of the MLEM algorithm [8] and it is reassuring that the PADE approach seems to share that feature. Yet, these small structures combined with low counts acquisition have posed a challenge for the MLEM model.…”

“…3c) but correlated statistical noise in the background, not present in the groundtruth, were also observed. This is an example of the images that were used to initialize the PADE v0 model, where its spots resolvability allow an efficient initialization of ω (see (8)). With 10 MLEM iterations (Fig.…”

“…We had observed that having ω i ≈ j φ * j,i , with φ * being the groundtruth, provided encouraging results. However, φ * being unknown, we opted to update ω with the previous iteration (see (8)) and use a low iteration MLEM to initialize the reconstruction (see ( 9)). While this approach makes it possible for ω to reach the desired value, it also means that the objective function is modified at each iteration.…”

“…As such, a better understanding of the benefits provided by ultra-fast TOF is of interest. For example, it was recently demonstrated that ultra-fast TOF resolution could provide a gain in spatial resolution by mitigating the blur induced by the detectors size [8]. Our goal is to investigate how ultra-fast TOF can be exploited in the PET reconstruction scheme to yield further gain in image quality and robustness.…”

Time of Flight in Perspective: Instrumental and Computational Aspects of Time Resolution in Positron Emission Tomography

“…The latter is traditionally done in an iterative way and requires knowledge of the distribution of activity and attenuation in image space. It has also been hypothesized that ~10 ps TOF information could be used in iterative reconstruction to improve the spatial resolution of high-resolution systems beyond the conventional limit imposed by the crystal pitch [ 224 ], though other effects, such as detector scatter may still limit such gains. Finally, due to the limited number of acquired events, some sort of noise suppression has to be included in the image generation process (either during or post reconstruction).…”