Research On The Inverse Problem Of Magneto-acoustic Concentration Tomography Of Magnetic Nanoparticles Based On Truncated Singular Value Decomposition

Magneto-acoustic magnetic nanoparticle concentration imaging(MACT-MI)is a high-resolution imaging technique,but current research is in its early stages.The proposed inverse problem algorithm has some problems such as boundary singularity and poor anti-noise performance,which makes it difficult to be applied to early breast cancer tumor recognition.Therefore,in order to further advance the research on the imaging inverse problem algorithm,the following studies are carried out in this paper:(1)A system matrix that relates the concentration distribution information of magnetic nano-particle swarm(MNPs)to ultrasonic signals is proposed,and a MNPs concentration reconstruction algorithm based on truncated singular value decomposition(TSVD)is proposed as well.The simulation model was established in COMSOL,and the magnetic field and sound field simulation data were substituted into the inverse problem algorithm based on TSVD for concentration reconstruction.The results show that the TSVD method can be used to reconstruct MNPs concentration accurately.(2)The influences of singular value number,signal-to-noise ratio and particle swarm radius on the reconstruction results were studied.The research results show that the proposed algorithm based on TSVD inverse problem can maximize the use of ultrasonic signals,and has a good reconstruction effect for MNPs with a small radius of 1mm.The reconstructed image can also be obtained with high resolution under the condition of low SNR,but the relevant errors and other data are not ideal.(3)In order to reduce the computational complexity and the reconstruction error,the modified truncated random singular value decomposition(MTRSVD)method was proposed to solve the system matrix,which is effective for solving the large-scale ill-posed problems in this paper.This paper has 29 figures,7 tables and 69 references.