Study On Energy Extraction Algorithm For Digital Measurement System Of In-beam PET

With the continuous development of particle detection technology,nowadays it is not only applied to basic disciplines such as particle physics and nuclear physics,but also widely applied to space science,astronomical observation,military,clean energy,environmental protection,geological exploration,contraband detection,agriculture,etc..Particle detection technology also has important applications in the field of nuclear medicine,not only for medical imaging,but also for radiotherapy of tumors.In particle detection,energy spectrum measurement is one of the important measurements.With the continuous improvement of detection accuracy,the scale of detectors and energy spectrum measurement systems has continued to grow,and the number of electronic channels also has continued to increase.The number of high-speed analog-to-digital conversion devices required for digital energy spectrum measurement systems is also increasing.The application of a large number of high-speed ADCs will bring the problems of system cost,data volume,power consumption,heat dissipation,and system complexity.This paper focuses on the energy signal processing requirements for digital measurement system of in-beam PET in heavy-ion cancer therapy device,and focuses on the study of energy extraction algorithms in order to reduce the ADC sampling rate as much as possible on the premise of obtaining higher energy resolution,so as to reducing the amount of data.And this paper focuses on the direct extraction pulse peaking algorithm,polynomial curve fitting algorithm,double exponential function curve fitting algorithm,and pulse area calculation algorithm.The output waveforms of the detector were obtained through an oscilloscope.The analysis modules were designed in MATLAB.Based on these algorithms,the data which obtained from six different lower sampling rates were analyzed and processed.The processing results are compared with the results of the direct extraction pulse peaking algorithm and pulse area calculation algorithm in high sampling rates.The validity of the algorithm is verified by discussing the relative error and calculating the energy resolution.Based on simulation experiments,the FPGA firmware module that implements the energy extraction algorithm is designed.In order to verify the feasibility and applicability of the energy extraction algorithm for digital measurement system of in-beam PET in heavy-ion cancer therapy device,the data sampled by the ADC were processed online in FPGA.Combined the simulation results with the experimental conditions and requirements,the 50 MSps sampling rate ADC was used in the in-beam PET principle prototype of heavy-ion cancer therapy device.The direct extraction pulse peaking algorithm and pulse area calculation algorithm were applied.The energy spectrum was measured for different detectors.The results of the energy spectrum statistics and the Flood Map statistics are consistent with the simulation results,which both show that the pulse area calculation algorithm has the advantages of high accuracy,good stability,and easy implementation.In the study of the algorithm,the influence of baseline drift and stacking distortion on energy resolution were considered.In the algorithm simulation stage and the experimental application stage,the average value is used to estimate the baseline value of the sampled waveform.And the baseline is subtracted from the pulse amplitude to achieve the purpose of baseline recovery.The pulse differential method is used to discard the pile up waveform data,which improves the energy resolution and achieves the expected goal.In the experimental application stage,the control variable method is used to conduct in-depth testing and discussion on ‘the important influence for the selection of the cumulative sum points in the pulse area algorithm on the energy resolution of the detector at a fixed sampling rate’.The test results show that the setting of an appropriate summation points is very important to improve the energy resolution of the detector.In addition,the paper briefly analyzes and discusses the front-end reading methods and circuits for weak signals.Influences of the front-end read circuit on the extraction of weak signals were analyzed and considered depending on the following factors,such as the selection of the amplifier,the structure of the preamp circuit,the selection of the feedback resistors and capacitors in the circuit,and the design of the protection ring on the PCB.It provides a reference for the hardware design and device selection of the energy spectrum measurement system.