State-of-the-art challenges and emerging technologies in radiation detection for nuclear medicine imaging: A review

Abstract

Single photon emission computed tomography (SPECT) and positron emission tomography (PET) are established medical imaging modalities that have been implemented for decades, but improvements in detector design and camera electronics are needed for advancement of both imaging technologies. Detectors are arguably the most important aspect of the systems. Similar to SPECT, PET typically relies on indirect conversion of gamma radiation via scintillators coupled with photosensors used to convert optical photons produced by the scintillator into an electrical signal. PET detectors are defined by their energy resolution, timing resolution, and spatial resolution, all of which affect and determine the image quality. Improvements in energy resolution have been shown by increasing the brightness of the scintillator utilizing materials like cerium bromide (CeBr3) or switching to a direct conversion detector, such as cadmium zinc telluride (CZT) or thallium bromide (TlBr). Timing resolution for PET is a focal point of the current research. Improving the timing resolution improves the signal-to-noise of the PET system and is integral to the implementation of time-of-flight PET. By utilizing novel configurations, such as side readouts on scintillators, timing resolution has been improved dramatically. Similarly, metascintillators, which use complex combinations for the scintillator material, have also shown improvements to the timing resolution. Additional research has focused on using Cherenkov light emission in scintillators to further improve the timing resolution. Other research is focused on using convolutional neural networks and other signal processing to enhance timing resolution. Lastly, aside from acollinearity and positron range, spatial resolution is impacted by the PET detector, therefore improving the intrinsic spatial resolution of the detector will allow for smaller features to be imaged. One method for improving the spatial resolution is to use unique configurations with layered scintillators. Additionally, monolithic scintillators have also been shown to have reduced spatial resolution. The future for both SPECT and PET image system advancement will depend on continued development of the detectors via many different pathways including materials, signal processing, physics, and novel configurations. In this review article, we will discuss challenges and emerging technologies for state-of-the-art radiation detectors utilized in PET and SPECT.

The detection instrumentation and geometric design of clinical PET scanner: towards better performance and broader clinical applications

“…A PET system's prowess in isolating true coincidence events from scattered coincidences augments the quality of reconstructed PET images since scattered events deteriorate image spatial resolution and contrast. Hence, refining energy resolution becomes imperative for PET, considering the elevated likelihood of scattering within human tissue [73]. Beyond detector energy resolution, other instrumental and geometric variables of the apparatus play a role in shaping this metric, like the scintillation detector's aperture, detector geometry, solid angle coverage, and acquisition mode (2D/3D); it is also influenced by the patient's physique and composition [74].…”

“…-Detection efficacy: Detector responsiveness to 511 keV γ photons hinges primarily on the detection medium's atomic number. Elevated sensitivity translates to better detector spatial resolution [73].…”

“…For this coupling arrangement, the intrinsic spatial resolution of the detector is primarily determined by the size of the individual pixels. Most current systems use pixelated scintillators with an intrinsic spatial resolution of around half the pixel size [73]. A significant challenge in manufacturing PET detectors with smaller scintillation crystal elements for better spatial resolution is that cutting, surface treating, and assembling the crystals to form arrays are complex and costly [155].…”

“…In this configuration, the interaction depth of the annihilation photon in the crystal is determined by the difference in the amount of light detected by the two photodetectors [14,69]. Results obtained using this configuration showed a DOI resolution of 2.9 ± 0.2 mm with a significant improvement in temporal resolution to 188 ± 32 ps [73]. A study was carried out based on a matrix coupling of SiPM-coupled LYSO scintillators in a double-sided readout (DSR) configuration, achieving an average DOI resolution of 4.1 mm, an energy resolution of 12.7%, and a coincidence time resolution of 353 ps [160].…”

“…The results obtained with this configuration are exceptional regarding time resolution, energy, and DOI resolution. The major drawback of these new detector configurations is the increase in the number of photodetectors and the complexity of the readout electronics [73].…”