Field-Based Spatial Self-Registration of Multicoil Hardware for B₀ Field Control

To develop and validate a magnetic field-based method for self-registration of multicoil (MC) shim hardware in MRI systems, enabling accurate B₀ field control despite nonreproducible hardware placement between calibration and experimental sessions.

An algorithm was designed to estimate rigid transformations between 3D field maps acquired during hardware calibration (Reference field) and subsequent experiments (Measured field). Hardware misplacement was determined by co-registration of the two fields within a defined region of interest. Simulations using Biot-Savart-derived MC basis fields and experimental validation with a 48-channel MC array at 3 T were used to quantify localization accuracy under varying signal-to-noise ratio (SNR) levels. A genetic algorithm optimized coil currents to generate a Reference field shape that yielded minimal self-registration error.

Simulations and measurements demonstrated that submillimeter (≤ 0.5 mm) and subdegree (≤ 0.5°) localization accuracy is achievable at practical SNR levels (≥ 5). Errors decreased with increasing SNR and field complexity. The optimized field produced by the genetic algorithm yielded the lowest mean translation error (0.20 mm) and rotation error (0.11°), consistent with theoretical predictions. Self-registration computation required less than 10 s per case.

The proposed field-based self-registration method enables rapid, accurate localization of MC shim hardware using only MR-acquired field maps, eliminating the need for external tracking hardware or repeated calibrations. This approach enhances reproducibility of MC field control and supports improved B₀ shimming performance in insert-based or repositionable MR systems.