International Journal on Magnetic Particle Imaging (IJMPI)
Not a member yet
555 research outputs found
Sort by
Recovering higher harmonics when increasing the frame rate in MPI
The frame rate of an MPI measurement can be increased by splitting the receive signal and reconstructing the split signals separately. Thus, motion artefacts can be reduced. Splitting the signal results in a decreased spectral resolution and a mismatch of higher harmonics. In this contribution, an approach for recovering the spectral resolution and higher harmonics is shown that is based on mirroring the split signals
The pitfalls of receive path calibration
In Magnetic Particle Imaging (MPI) and magnetic particle spectroscopy (MPS) magnetic nanoparticles (MNPs) are exposed to static and dynamic fields. These cause a dynamic magnetization response that is typically measured with inductive coils. The signal acquisition generally occurs in parallel with the excitation. This has the consequence that the excitation field couples into each receive path. The feed-through signal is commonly dampened by advanced passive filtering, at the cost of a distorted particle signal. Consequently, the measurement signals of different MPI or MPS devices will differ, even if the underlying magnetization response of the MNPs is the same. Receive path calibration can be used to address this issue by reverting these distortions and transforming the signal into a device independent domain. The authors of this abstract studied a general calibration procedure for multi-channel, non-orthogonal and non-homogeneous receive coils along with an analytical calibration model. Furthermore, method and model uncertainties were investigated and a systematic model error that had not been accounted for in previous calibration methods has been identified. This systematic model error could be attributed to the approximation of the mutual inductance between receive and calibration coil and it becomes non-negligible in experimental setups with small inductive receivers. Suggestionswere made for estimating and reducing its influence. Finally, the method was used to calibrate the receive path of an MPS system and of a multi-channel, non-orthogonal MPI receive coil setup
A Development of 3D navigation system for micro-nano robot based on a Magnetic Particle Imaging system
The 3D navigation system of micro-nano robot based on magnetic particle imaging is a promising technique to be used for targeted drug delivery. Using the magnetic actuator by manipulating field free point (FFP) and non-invasive molecular imaging method such magnetic particle imaging, the device can reduce the harm in treatment. In this paper, we used closed-loop control to steer a micro robot to a desired position by a 3D FFP magnetic force based system. A custom-built MPI system with a bore size of 90 mm that can generate a gradient as high as 4T/m/µ0 was used to verify the control system in three dimensions
Recovery of fundamental frequency component in magnetic particle imaging Using an attention-based neural network
Magnetic particle imaging (MPI) is a rapidly developing medical imaging modality, which uses the nonlinear response of superparamagnetic iron oxide nanoparticles to the applied magnetic field to image their spatial distribution. Due to the direct feedthrough of excitation signals, the existing MPI systems directly filter out the fundamental frequency component of the received signal, resulting in the loss of first harmonic information. In this work, we proposed a deep learning (DL) method adopting self-attention mechanism, which can effectively recover fundamental frequency component of the signals in the presence of background noise. At the same time, our method deals with two-dimensional time-frequency spectrum obtaining by piecewise Fourier transform from the time domain signals. The performance of our method is analyzed via simulation experiments, which show that our method can effectively recover first harmonic information and obtain high quality MPI reconstructed images
A design of a low-power open-side MPI scanner Using inverted spliced permanent magnets
Magnetic particle imaging (MPI) is a promising imaging modality with high sensitivity, high resolution and no depth limitation. Scaled MPI scanners are developed to satisfy the demand for imaging larger imaging objects. Considering the ultimate goal of whole-body scanning, open-side MPI scanner is more suitable than traditional closed-bore MPI scanner. A fully electrically driven open-side MPI scanner have been built, enabling rotating and translating field free line (FFL) in three dimensions. However, the high power required to run the system may become a challenge when trying to scale it up. Here, we present a design of an open-side MPI device using inverted spliced permanent magnets. The design eliminates the need for power suppliers to generate and move the FFL, reducing the cooling requirements of the system. Compared to conventional solution based on permanent magnets, our solution can generate a more uniform FFL, which is essential for accurate imaging
2D Projection Imaging with a Single-Sided FFL Magnetic Particle Imaging Scanner
There have been several MPI scanner designs with the promise of useful clinical imaging. One such promising design is the so called single-sided scanner, which limits all hardware to a single side of the device. This provides the scanner with an unrestricted imaging volume, but at the cost of finite imaging depth and inhomogeneous magnetic fields. Our group has designed such a scanner that utilizes an FFL for spatial encoding, which differs from the original single-sided scanner which utilizes an FFP. In this work we demonstrate the first two dimensional images obtained using a single-sided FFL MPI scanner
Flexible Selection Field Generation using Iron Core Coil Arrays
Many different concepts for selection-field generators have been introduced for Magnetic Particle Imaging so far. Inthis work, the field generation characteristics of an optimized iron core selection-field generator consisting of twocoil arrays with a total of 18 coils are presented. Due to the high number of degrees of freedom, a wide variety offield configurations are possible. The setup allows the generation of arbitrarily shaped fields, including the standardMagnetic Particle Imaging fields such as field-free points and field-free lines. In this work, field measurements arepresented and the current calculation method for generating a specific field configuration is discussed
Towards a Fully Integrated Preclinical Field-Free Line MPI Scanner
As a radiation-free, highly sensitive and high resolution imaging technique, magnetic particle imaging promisesto provide deep insights into biological processes of living specimen. A permanent magnet-based, field-free line(FFL) scanner for the scope of molecular imaging on cell level or in small animals is presented. The scanneris based on two rotatable discretized Halbach dipole rings creating a FFL between them, that is orthogonalto their common axis. Using a solenoid coil the FFL can be translated in the imaging plane. The combinedrotation and translation of the FFL results in a native 2D image acquisition that can be extended to 3D by amoving table approach. Due to the high gradient of 5 Tm????1, the addition of a focus field is needed to extend thefield of view to the full 40mm bore size. Both the drive and focus field are generated by superimposed currentsin a single coil. The currents are driven by two separate sources which are decoupled by a protection filter.Signal acquisition is realized with a gradiometric pickup coil and first phantom experiments suggest an in-planespatial resolution smaller than 0.7mm. The hardware and software concept of the scanner will be presented, aswell as the iterative improvements and design decisions that were made based on inter-dependencies betweenthe individual components. In conclusion, a platform for a scanner will be shown that is aimed directly atmolecular imaging
Parallel MPI image reconstructions in GPU using CUDA
This work shows that it is possible to obtain faster MPI image reconstructions by implementing the algorithms in parallel in Graphics Processing Units (GPUs) using NVIDIA’s CUDA (Compute Unified Device Architecture). While the parallel Kaczmarz’s algorithm was slower than its serial version running in the Central Processing Unit (CPU), the parallel version of the Conjugate Gradient Normal Residual (CGNR) algorithm was about 58 times faster than its serial implementation, and about 10 times faster than the serial implementation of Kaczmarz’s
X-Space Image Reconstruction for Lissajous Trajectory Using Multidimensional Image Tensor
The tensor-based theory of multidimensional x-space MPI provides useful insight into MPI image reconstruction. Using this theory, it was shown that x-space MPI images with isotropic resolution can be achieved by scanning in two orthogonal directions separately and combining the resulting images. In this work, we propose an x-space image reconstruction that resolves the multidimensional image tensor, allowing us to reconstruct the isotropic MPI image for the Lissajous trajectory. The proposed method takes advantage of the self-crossing property of the Lissajous trajectory