Performance Of SiPM Based Semi-monolithic Scintillator PET Detectors

Two depth encoding positron emission tomography(PET)detector modules using semi-monolithic scintillation crystal single-ended readout by silicon photomultiplier(SiPM)arrays were built,and their performance was measured.The first semi-monolithic scintillation detector consists of 11 polished LYSO slices measuring 1×11.6×10 mm3.The slices were glued together with the enhanced specular reflector(ESR)in between and outside of the slices.The bottom surface of the slices was coupled to a 4×4 SiPM array.No reflector was used on the top surface and two sides of the slices.The signals of the 4×4 SiPM array are grouped along rows and columns separately into 8 signals.Four row signals are used to identify the slices according to the center of the gravity(COG)of the scintillation photon distribution in the pixelated direction.Both the least-square minimization(LS)and the maximum likelihood(ML)positioning algorithms were developed to simultaneously estimate the y-(monolithic direction)and z-(depth direction)positions by use of four column signals.The detector was measured with 1 mm sampling interval in both the y-and z-directions with electronic collimation by using a 22Na point source and a 1 × 1 ×20 mm3 LYSO crystal reference detector.An average slice based energy resolution of 14.9%was obtained.All slices were clearly resolved.Edge effects existing in monolithic detectors were clearly observed.Both the LS and ML positioning algorithms improved the spatial resolution at the crystal edges as compared with the simple COG method.The mean absolute error(MAE)which is defined as the probability-weighted mean of the absolute value of the positioning error was used to evaluate the positioning accuracy.Performance of the two methods is similar.An average MAE spatial resolution of?1.15 mm was obtained in both the y-and z-directions.The configuration of the second semi-monolithic scintillation detector is similar to the first one.It consists of 11 longer polished LYSO slices measuring 1”37.6”10 mm3.On the top surface and two sides the slices,two surface treatment methods were tested.One is no use of reflector nor black paint,and the other is finishing with black paint and then wrapping with black tape.The bottom surface of the slices was coupled a 4”12 SiPM array.SiPM signals were grouped along rows and columns separately into 16 signals.Four rows are used to identify slices,and 12 columns are used to estimate the y-and z-positions.In order to develop positioning methods accessible in front-end circuits,the COG and the squared COG method were tested for y-position estimation.The standard deviation of column signals,the sum of squared column signals,and the ratio of maximum to the total of column signals were tested for z-position estimation.The second detector was measured in the same way as the first one did.All slices were clearly resolved,regardless of surface treatment methods.Positioning results show that the squared COG method is superior to the COG method and that all the three depth positioning methods can provide high resolutions enough for the parallax error correction.Compared with the no black paint detector,the average spatial resolution of the black paint detector is significantly improved.It is improved from an FWHM of 1.97 mm to an FWHM of 1.20 mm in the y-direction using the squared COG method,and from an FWHM of 2.92 mm to an FWHM of 1.72 mm by using the standard deviation method in the z-direction.But,the slice based average energy resolution degrades from 15.8%to 24.9%.The timing resolution of the entire detector module also deteriorates from 596 ps to 788 ps.