Hybrid x-space: a new approach for MPI reconstruction

Abstract

Magnetic particle imaging (MPI) is a new medical imaging technique capable of recovering the distribution of superparamagnetic particles from their measured induced signals. In literature there are two main MPI reconstruction techniques: measurement-based (MB) and x-space (XS). The MB method is expensive because it requires a long calibration procedure as well as a reconstruction phase that can be numerically costly. On the other side, the XS method is simpler than MB but the exact knowledge of the field free point (FFP) motion is essential for its implementation. Our simulation work focuses on the implementation of a new approach for MPI reconstruction: it is called hybrid x-space (HXS), representing a combination of the previous methods. Specifically, our approach is based on XS reconstruction because it requires the knowledge of the FFP position and velocity at each time instant. The difference with respect to the original XS formulation is how the FFP velocity is computed: we estimate it from the experimental measurements of the calibration scans, typical of the MB approach. Moreover, a compressive sensing technique is applied in order to reduce the calibration time, setting a fewer number of sampling positions. Simulations highlight that HXS and XS methods give similar results. Furthermore, an appropriate use of compressive sensing is crucial for obtaining a good balance between time reduction and reconstructed image quality. Our proposal is suitable for open geometry configurations of human size devices, where incidental factors could make the currents, the fields and the FFP trajectory irregular.

A novel software framework for magnetic particle imaging reconstruction

“…First, the reconstruction module integrates the classical system matrix‐based and the x‐space reconstruction methods. More advanced reconstruction methods, such as the sparse system matrix reconstruction 23 and the hybrid x‐space reconstruction methods 19 can be integrated into MPIRF. Second, the main parameters of the MPI scanner were used in MPIRF.…”

“…14 Many MPI image-reconstruction methods have been reported in the past decade; they can be mainly divided into system matrix-based 15,16 and x-space methods. [17][18][19] The reconstruction framework integrates the reconstruction methods, reduces the negative computer computing power of new reconstruction methods, and simplifies the deployment of the new method in the application environment, thereby reducing the application complexity of the new method, accelerating new method development, and promoting MPI development in clinical applications. However, there is no unified software framework for integrating the two MPI image reconstruction methods.…”

Magnetic particle imaging: tracer development and the biomedical applications of a radiation-free, sensitive, and quantitative imaging modality

“…Pre-clinical system Please do not adjust margins Please do not adjust margins specifications are slowly improving as researchers have been highly active in designing improved software for image acquisition and reconstruction. 35,37,39,40,[365][366][367][368][369][370][371][372][373] For example, Gladiss et al recently demonstrated the advantages of a hybrid system matrix for efficient calibration and image reconstruction. 374 To scale-up MPI hardware to human-size, there are several notable safety considerations and ergonomic challenges that must be overcome.…”

Applications of PDEs inpainting to magnetic particle imaging and corneal topography

“…On the other hand, XS works only for uniformly spatial drive fields. In recent years, a different Hybrid XS (HXS) approach, combining both the MB and XS, was also proposed in [55,56]. An interesting overview of all methods techniques adopted in MPI, together with their preclinical investigations, can be found in [33].…”

“…The parameter values used in the corresponding numerical algorithms are listed in Table 1 (Step 2). The resulting images u x and u y was firstly processed by a Wiener's deconvolution (Step 3) and successively merged pixel by pixel into the final image u by a geometrical mean, that is u(i, j) = u x (i, j)u y (i, j), as done in [56] (Step 4). At this last step a further processing by a high-pass filter was performed.…”