Research On Performance Optimization Strategy Of Fluorescence Molecular Tomography

As an imaging method,fluorescence molecular tomography(FMT)is aimed to reconstruct three-dimensional fluorescence mass distribution of biological tissue,which has broad application prospects in the field of medical image.Due to the strong scattering of light in biological tissue,the limited of obtained surface fluorescence data and noise contained etc,source reconstruction of FMT is an ill-posed problem and the reconsruction quality should be improved.In this paper,some performance optimization strategies was develpoed in FMT research by optimizing the excitation point setting and adaptively obtaining the threshold value from the recovered images.The specific research work is as follows:1)Imaging system of fluorescence molecular tomography with spiral excitation.Existing FMT imaging system usually was based on multiple point excitation,and the imaging results are closely related to excitation point setting.When excitation point setting is inconsistent with the position of fluorescent target,reconstruction results of circle excitation mode have great errors.Due to small excited area,the array excitation mode has a long time to collect data.In order to solve the problem of incomplete excitation of fluorescence target,this paper presents a spiral multi-point excitation fluorescence molecular tomography strategy.In this spiral excitation mode,the excitation points distribute around the imaging target in spiral structure.By expanding the excitation points distribution range,the fluorescence target can be completely excited to guarantee the accuracy of data collection,even though the specific location of fluorescence target is unknown.Through the heterogeneous phantom single source and dual source experiments,by comparing the reconstruction results of circle excitation,array excitation and spiral excitation,the comparing results show that the collecting data of spiral excitation method is more accurate and reconstruction error is smaller.2)Adaptive threshold method for recovered images of FMT.There are many pseudo images in recovered images of FMT.To avoid some misleading information,the recovered results are displayed by selecting nodes with a threshold.In general,the threshold was set by artificial,so the results were not accurate and objective enough.In order to automatically select the threshold according to the results of reconstruction,this paper presents a post-processing strategy for recovered images of FMT.A threshold value is adaptively obtained from the recovered images without external interference,which is objective because it is extracted from the reconstructed result.Through the digital simulation and physical phantom experiments,it is can be seen that the images of this adaptive threshold method are clearer than those with no post-processing,which is easy to evalute the results.The full width at half-maximum and contrast-to-noise ratio are then utilized to further verify the effectiveness of the post-processing method,being capable of removing spurious information from the original images,thus bringing convenience to users.