Sub-millimeter precise photon interaction position determination in large monolithic scintillators via convolutional neural network algorithms

Abstract

In this work, we present the development and application of a convolutional neural network (CNN)based algorithm to precisely determine the interaction position of γ-quanta in large monolithic scintillators. Those are used as an absorber component of a Compton camera (CC) system under development for ion beam range verification via prompt-gamma imaging. We examined two scintillation crystals: LaBr 3 :Ce and CeBr 3 . Each crystal had dimensions of 50.8 mm × 50.8 mm × 30 mm and was coupled to a 64-fold segmented multi-anode photomultiplier tube (PMT) with an 8 × 8 pixel arrangement. We determined the spatial resolution for three photon energies of 662, 1.17 and 1.33 MeV obtained from 2D detector scans with tightly collimated 137 Cs and 60 Co photon sources. With the new algorithm we achieved a spatial resolution for the CeBr3 crystal below 1.11(8) mm and below 0.98(7) mm for the LaBr3:Ce detector for all investigated energies between 662 keV and 1.33 MeV. We thereby improved the performance by more than a factor of 2.5 compared to the previously used categorical average pattern algorithm, which is a variation of the well-established k-nearest neighbor algorithm. The trained CNN has a low memory footprint and enables the reconstruction of up to 10 4 events per second with only one GPU. Those improvements are crucial on the way to future clinical in vivo applicability of the CC for ion beam range verification.

A new detector concept based on the prompt gamma radiation analysis for In vivo boron monitoring in BNCT

“…Thus, we have proposed to use monolithic lanthanum bromide crystals with a silicon photomultipliers light readout with an anti-Compton shield. In the first prototype we used 2″ × 2″ crystal read out by 62 SiMPs, which may enable the determination of a place of γ quantum interaction with a few millimeters of precision using neural network algorithms ( 21 ) . This possibility, together with the rejection of Compton scattered γ quanta background may significantly decrease the background and increase the signal-to-background ratio for our PGRA-SPECT system being under construction.…”

Gamma Ray Emission Imaging in the Medical and Nuclear Safeguards Fields

“…This minimizes detection dead time in case of high count rates, which are, e.g. typical in nuclear imaging with very short-lived isotopes such as 82 Rb (halflife 76 s). There is in principle no one-to-one relationship between fast signals and good timing resolution, but in practice this is most often the case.…”

“…Obtaining the 3D position information in such detectors requires an elaborate calibration procedure. Most recently, neural networks are used to determine the position of individual events [82].…”

Joint estimation of interaction position and energy deposition in semiconductor SPECT imaging sensors using fully connected neural network

“…With this flexible reconstruction method, welldesigned scintillator modules are modeled to test their intrinsic spatial resolution [16,17]. Similarly, submillimeter accuracy positioning can be achieved with fine beam scanning of a monolithic scintillation crystal [18]. Other applications include the time-of-flight prediction from waveforms [19] and signal discrimination for CMOS probes [20].…”