Genetic algorithms for MRI magnet design (original) (raw)
Related papers
A Comparison of Two Design Methods for MRI Magnets
IEEE Transactions on Appiled Superconductivity, 2004
Designs of magnetic resonance imaging (MRI) main magnets obtained from both a functional method and a genetic algorithm method have been compared. While most features in the two approaches are similar, there are several important differences. The functional method leads to fewer coil bundles and a reduced total current, i.e., total ampere turns, (e.g., 6-8 MA) that can be as much as 70% of the total current found with the genetic analysis. While the conclusion about stress is that it is a sensitive function of the choice of wire current density, the designs found with the functional method have a larger hoop stress than that of the genetic design, which may require new or refined manufacturing techniques. Furthermore, the functional approach requires much less computing power (i.e., a personal computer is quite sufficient) while the genetic algorithm method in general demands massively parallel computing techniques. However, in order to search for the optimal magnetic resonance design at a given field strength, it is likely that a combination of these two methods will lead to the best results.
2005
This paper introduces the design of superconductive magnet configurations in Magnetic Resonance Imaging (MRI) systems as a challenging real-world problem for Evolutionary Algorithms (EAs). Analysis of the problem structure is conducted using a general statistical method, which could be easily applied to other problems. The results suggest that the problem is highly multimodal and likely to present a significant challenge for many algorithms. Through a series of preliminary experiments, a continuous Estimation of Distribution Algorithm (EDA) is shown to be able to generate promising designs with a small computational effort. The importance of utilizing problem-specific knowledge and the ability of an algorithm to capture dependencies in solving complex real-world problems is also highlighted.
Asymmetric MRI Magnet Design Using a Hybrid Numerical Method
Journal of Magnetic Resonance, 1999
This paper describes a hybrid numerical method for the design of asymmetric magnetic resonance imaging magnet systems. The problem is formulated as a field synthesis and the desired current density on the surface of a cylinder is first calculated by solving a Fredholm equation of the first kind. Nonlinear optimization methods are then invoked to fit practical magnet coils to the desired current density. The field calculations are performed using a semi-analytical method. A new type of asymmetric magnet is proposed in this work. The asymmetric MRI magnet allows the diameter spherical imaging volume to be positioned close to one end of the magnet. The main advantages of making the magnet asymmetric include the potential to reduce the perception of claustrophobia for the patient, better access to the patient by attending physicians, and the potential for reduced peripheral nerve stimulation due to the gradient coil configuration. The results highlight that the method can be used to obtain an asymmetric MRI magnet structure and a very homogeneous magnetic field over the central imaging volume in clinical systems of approximately 1.2 m in length. Unshielded designs are the focus of this work. This method is flexible and may be applied to magnets of other geometries.
1993
The ROXIE software program package has been developed for the design of the superconducting magnets for the LHC at CERN. The software is used as an approach towards the integrated design of superconducting magnets including feature-based coil geometry creation, conceptual design using genetic algorithms, optimization of the coil and iron cross-sections using a reduced vector-potential formulation, 3-D coil end geometry and field optimization using deterministic vector-optimization techniques, tolerance analysis, production of drawings by means of a DXF interface, end-spacer design with interfaces to CAD-CAM for the CNC machining of these pieces, and the tracing of manufacturing errors using field quality measurements. This paper gives an overview of the methods appied in the ROXIE program.
2003
In this paper we are concerned with the design of a small low-cost, low-field multipolar magnet for Magnetic Resonance Imaging with a high field uniformity. By introducing appropriate variables, the considered design problem is converted into a global optimization one. This latter problem is solved by means of a new derivative free global optimization method which is a distributed multi-start type algorithm controlled by means of a simulated annealing criterion. In particular, the proposed method employs, as local search engine, a derivative free procedure. Under reasonable assumptions, we prove that this local algorithm is attracted by global minimum points. Additionally, we show that the simulated annealing strategy is able to produce a suitable starting point in a finite number of steps with probability one.
Coil optimization for MRI by conjugate gradient descent
Magnetic Resonance in Medicine, 1991
A flexible iterative algorithm is presented for optimizing gradient and radio frequency coils for MRI. It is based on a model using discrete current elements and direct Biot-Savart calculation of the fields. An error function is defined over a region of interest (R01) and is minimized by conjugate gradient descent. The choice of error function allows optimization of the field uniformity, the inductance, and the efficiency of the coil in any combination. Neither the coil nor the ROI is restricted to any particular geometry. A 40turn cylindrical z-gradient coil of radius a and length 4a, designed for a R01 of radius 0.7a and length 2a has an average error in the gradient fields generated of 0.85%, an inductance of 0.014a mH/cm, and an efficiency of 6.6W2 Gcm/A. A 16-turn birdcagelike RF coil of radius 5 cm, designed for a ROI of radius 4 cm has an average error of 0.79%.
Racetrack Coils for Dedicated MRI Magnets
IEEE Transactions on Applied Superconductivity, 2010
The need to optimize the magnet shape, size and region of homogeneity with respect to the anatomy of the patient is particularly strong in dedicated MRI magnet design: cost and weight are often determining factors for success in the market. An elongated coil geometry could be useful in cases where the scanner is designed for imaging non axisymmetric parts of the human body (e.g. extremities): in this paper we present a fast design method based on Linear Programming (LP), in which racetrack coils are used to optimize the magnetic field homogeneity over an ellipsoidal region, together with some results.
Design of Gradient Coil for Magnetic Resonance Imaging Applying Particle-Swarm Optimization
IEEE Transactions on Magnetics, 2011
Designing a gradient coil for magnetic resonance imaging (MRI) is an electromagnetic inverse problem often formulated as a constrained optimization, which has been successfully solved by inverse boundary element methods. The constant search for new coil features and improved performance has highlighted the need of employing more versatile optimization techniques capable of dealing with the new requirements. In this paper, the solution of linear and nonlinear optimization problems using particle-swarm optimization (PSO) algorithms is presented. Examples of coil designed using this heuristic method are shown, including a comparison to solutions provided by conventional optimization approaches. Numerical experiments reveal that the application of PSO for the solution of inverse boundary element problems for coil design is a computationally efficient algorithm that is capable of handling nonlinear problems and that offers fast convergence, especially for those symmetric coil geometries where the computational effort can be drastically reduced by using suitable dimensionality-reduction techniques.
Resistive homogeneous MRI magnet design by matrix subset selection
Magnetic Resonance in Medicine, 1999
A new technique for designing resistive homogeneous multicoil magnets for magnetic resonance imaging (MRI) is presented. A linearly independent subset of coils is chosen from a user-defined feasible set using an efficient numerical algorithm. The coil currents are calculated using a linear least squares algorithm to minimize the deviation of the actual magnetic field from the target field. The solutions are converted to practical coils by rounding the currents to integer ratios, selecting the wire gauge, and optimizing the coil cross-sections. To illustrate the technique, a new design of a short, homogeneous MRI magnet suitable for low-field human torso imaging is presented. Magnets that satisfy other constraints on access and field uniformity can also be designed. Compared with conventional techniques that employ harmonic expansions, this technique is flexible, simple to implement, and numerically efficient. Magn