Design And Development Of In Vivo EPR Measurement Software

Nuclear and radiation events may cause ionizing radiation to a certain range of people,resulting in a certain degree of bodily harm.After personnel are exposed to radiation,rapid and accurate estimation of radiation dose is crucial for determining treatment plans and achieving maximum utilization of medical resources.The in vivo electron paramagnetic resonance(EPR)method can be used for on-site,rapid,and non-invasive detection of radiation dose to injured personnel.This article is based on the working principle and process of the existing in vivo EPR measurement system.Starting from two spectral evaluation methods,interactive evaluation and intelligent evaluation,an in vivo EPR measurement software that can be used for spectral collection,screening,and processing is designed and developed to complement the system.The software part of the control and data unit of the in vivo EPR measurement system is improved.The main work of this article is as follows:(1)Firstly,this paper analyzes the working principle and system composition of the in vivo EPR measurement system,determines that the control and data unit of the system needs to be further improved,Based on the working principle of the in vivo EPR measurement system,the requirements analysis and overall design of the software are carried out.The functional requirements of the software are divided into three modules: EPR spectrum acquisition,EPR spectrum screening,and EPR spectrum data processing.(2)The application of machine learning in EPR spectral evaluation was studied.Using Support Vector Machine(SVM)method,a method for automatic classification and recognition of EPR spectra was established.Using Genetic Algorithm Back Propagation(GA-BP)method,a method for predicting the radiation dose to the injured was established,which can automatically identify radiation induced signals and predict their radiation dose.(3)Based on the functional requirements and workflow of the software,a detailed design was carried out on the functional modules of the in vivo EPR measurement software.Functional testing and experimental verification were conducted on the software’s functions and the application of machine learning in EPR spectrum evaluation.The results showed that the software can meet the needs of system functions and the feasibility of machine learning in EPR spectrum evaluation.