Dual Mode Image Reconstruction Of Glioma Based On Random Enhancement Adaptive Algorithm With Small Support Set

The fusion of CT modality and optical molecular modality can detect the deep brain glioma in the early stage of the disease,and can accurately locate and quantitatively analyze the tumor.However,the dual modality three-dimensional tomographic reconstruction of glioma is a pathological problem with serious ill posed nature.In this paper,we use the random enhancement adaptive algorithm based on small support set to reconstruct the brain tumor with dual modality,and apply it to medical research.The main work of this article is as follows:A random enhancement adaptive algorithm based on small support set is proposed to reconstruct the light source.The random enhancement adaptive algorithm based on small support set is a greedy pursuit algorithm.The random strategy is used to select unrelated atoms,and the new element which belongs to the last iteration process is abandoned in advance,and the convergence of the algorithm is improved.A heterogeneous phantom model is used to verify the feasibility of the proposed algorithm in bioluminescent tomography.The proposed algorithm can effectively divide the bioluminescent signal from the background,and the reconstruction efficiency is twice as much as that of the random enhancement adaptive algorithm.In addition,the dual mode data of in vivo mice are collected and preprocessed,the random enhancement adaptive algorithm based on small support set is used to reconstruct the glioma of the tumor bearing mice.Finally,the three dimensional tomographic reconstruction of optical molecular images was applied to the experimental study: The effectiveness of temozolomide in the treatment of in situ brain tumor was evaluated by bioluminescence tomography based onmultimodality imaging system;Matrix metalloproteinase activated NIR fluorescence imaging and image guided glioma resection were performed.