The path to the “ideal” brain PET imager: The race is on, the role for TOF PET

Majewski, Stanislaw (2020) The path to the “ideal” brain PET imager: The race is on, the role for TOF PET. Il nuovo cimento C, 43 (1). pp. 1-35. ISSN 1826-9885

Text ncc12018.pdf - Published Version Download (2MB)

Abstract

With the efforts under way to improve spatial resolution of the revolutionary Explorer family of imagers, the acute need to develop dedicated imagers for breast, prostate, heart, etc. may slowly disappear, except for some specialized cases in treatment guidance and monitoring, for example in proton therapy. It is in fact happening already. Part of the reason is the high cost of the dedicated systems but also an intriguing emerging opportunity that long axial length PET scanners can be equipped with magnifying inserts that can locally boost the resolution, as per the so-called virtual pinhole concept by Yuan-Chuan Tai from WashU, also called Zoom-in PET. However, the exception are the brain imaging scanners. The special geometry of the optimal helmet type designs for imaging of the brain still gives the opportunity to the brain PET imager developers to compete for the “best” system. We all want to produce good quality dynamic molecular PET brain images at low injected radiation doses (and... low cost). Several designs are being proposed as well as being built at this time in many places around the world. These designs mostly fall in two categories: 1) the mini-Explorer cylindrical type or 2) the compact helmet type, both with large angular brain coverage assuring high sensitivity. Due to the compact sizes of the helmet-type systems, in order to substantially benefit from the improved TOF performance, one needs to achieve better than 100 ps FWHM timing performance. In fact, 50 ps FWHM would be a very nice goal. Several groups are working on such concepts. In this race, any new ideas from the expert instrumentation community (not only the medical one) are highly encouraged, as a great impact is expected on brain imaging once such high-performance but also dissemination-ready (i.e., robust and economical) designs are developed. Ideally, the brain imagers of the next generation will have high sensitivity and high spatial resolution approaching the predicted physical limit (due to positron range plus non-collinearity of the two emitted annihilation photons), limited to about 1 mm FWHM. Interestingly, there is a known connection between spatial resolution and sensitivity in detecting small lesions or structures, through the Partial Volume Effect (PVE). The adversarial effect of poor resolution on the detection of small structures is the blurring of the signal with the background. Inversely, if there is not enough statistics (detected/recorded events) per reconstruction voxel, even the best spatial resolution will not bring the tomographic uptake signal above the noisy background.

Item Type:	Article
Subjects:	500 Scienze naturali e Matematica > 530 Fisica
Depositing User:	Marina Spanti
Date Deposited:	29 Jan 2021 15:26
Last Modified:	29 Jan 2021 15:26
URI:	http://eprints.bice.rm.cnr.it/id/eprint/20732

Actions (login required)

View Item