Abstract
PURPOSE: The increasing prevalence of orthopedic metallic implants necessitates optimization of MRI methods to monitor surrounding tissues and identify complications. The emergence of contemporary mid- and low-field systems as promising platforms for imaging around metal requires development of new protocols optimized for these field strengths. Motivated by these, we develop a simulation framework that predicts MRI performance near metallic implants, considering B0 field strength and multi-spectral imaging (MSI) parameters. METHODS: A simulation framework for imaging near metal incorporating 3D implant and 3D human anatomical models, and imaging and sequence parameters is developed. MSI acquisitions are simulated based on susceptibility-induced field shifts. Validation is performed using phantom and in vivo experiments at 0.55 and 3 T, comparing simulated and experimental images for artifact shape and SNR. Performance is demonstrated by evaluating impacts of field strength, spectral bins, RF and readout bandwidths, and implant materials. RESULTS: Simulations accurately reproduced experimental imaging performance, with extent and patterns of metal artifacts closely matching experimental observations at both 0.55 and 3 T, while correctly predicting increased SNR and artifact size with increasing field strength. The framework demonstrated the impact of field strength, spectral bins, RF and readout bandwidths, and implant material on metal artifacts and image quality. CONCLUSION: This open-source simulation framework effectively predicts MRI performance near metallic implants. This can facilitate protocol optimization and exploration of imaging parameters without extensive/costly in vivo experiments. This tool can enable researchers to predict imaging performance for various implants across different field strengths, potentially accelerating development of optimized protocols for clinical use.