Abstract
Parallax error is a common issue in high-resolution preclinical positron emission tomography (PET) scanners as well as in clinical scanners that have a long axial field of view (FOV), which increases estimation uncertainty of the annihilation position and therefore degrades the spatial resolution. A way to address this issue is depth-of-interaction (DOI) estimation. In this work we propose two machine learning-based algorithms, a dense and a convolutional neural network (NN), as well as a multiple linear regression (MLR)-based method to estimate DOI in depolished PET detector arrays with single-sided readout. The algorithms were tested on an 8× 8 array of 1.53× 1.53× 15 mm3 crystals and a 4× 4 array of 3.1× 3.1× 15 mm3 crystals, both made of Ce:LYSO scintillators and coupled to a 4× 4 array of 3× 3 mm3 silicon photomultipliers (SiPMs). Using the conventional linear DOI estimation method resulted in an average DOI resolution of 3.76 mm and 3.51 mm FWHM for the 8× 8 and the 4× 4 arrays, respectively. Application of MLR outperformed the conventional method with average DOI resolutions of 3.25 mm and 3.33 mm FWHM, respectively. Using the machine learning approaches further improved the DOI resolution, to an average DOI resolution of 2.99 mm and 3.14 mm FWHM, respectively, and additionally improved the uniformity of the DOI resolution in both arrays. Lastly, preliminary results obtained by using only a section of the crystal array for training showed that the NN-based methods could be used to reduce the number of calibration steps required for each detector array.
Highlights
Image quality in PET is directly affected by the number of detected photons
On the side opposite to the silicon photomultipliers (SiPMs), each detector module was equipped with a 70 μm-thick enhanced specular reflector (ESR) film (3 M, USA) placed on top of a 1 mm-thick glass light guide used for redirection and sharing the scintillation light among the crystals in the array
Neural network-based algorithms To account for nonlinearities in light distribution response to DOI, we have investigated two machine learning approaches for DOI estimation, a dense neural network (NN) (DNN) and a convolutional NN (CNN)
Summary
Image quality in PET is directly affected by the number of detected photons. All commercial PET scanners and most research scanners today use scintillation crystals for gamma detection. The interaction probability of an annihilation photon within the scintillator depends on multiple factors, one of which is the travelling distance within the material. Increasing the length of the crystals is a simple way to increase the detection probability and enhance the scanner sensitivity. The uncertainty in estimation of the interaction point inside the crystal leads to a resolution-degrading effect known as the parallax error. Both of these effects can be mitigated by depth-of-interaction (DOI) estimation
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
