Accelerated imaging of metallic implants using model-based nonlinear reconstruction.

Area of Science:

• Magnetic Resonance Imaging (MRI)
• Medical Imaging Physics
• Biomedical Engineering

Background:

• Metal implants cause significant magnetic field distortions, complicating MRI near these devices.
• Multi-spectral imaging (MSI) acquires data at multiple frequencies to mitigate metal artifacts but is computationally intensive.
• Accelerating MSI is crucial for clinical feasibility and patient comfort during MRI scans.

Purpose of the Study:

• To develop and evaluate a model-based reconstruction technique for accelerating MSI near metallic implants.
• To exploit spectral correlations within MSI data for improved image reconstruction.
• To assess the accuracy and performance of the proposed method in simulations and in vivo.

Main Methods:

• A novel signal model was developed to capture the correlation between spectral bins in MSI data.
• Model-based reconstruction was implemented, enforcing the signal model during image reconstruction.
• The method was evaluated using simulations and in vivo MRI data from 6 subjects at a 17.4x acceleration factor.
• Performance was compared against model-free parallel imaging and compressed sensing (PI & CS) using nRMSE and SSIM metrics.

Main Results:

• The proposed signal model accurately and compactly represents MSI signal profiles across spectral bins.
• Model-based reconstruction significantly improved image quality compared to model-free PI & CS.
• Statistically significant improvements (P < 0.05) in nRMSE and SSIM were observed at 17.4x acceleration in vivo.
• Simulations and in vivo experiments confirmed the method's effectiveness.

Conclusions:

• Model-based reconstruction is a feasible approach to accelerate MSI techniques.
• This method enables faster MRI acquisition in the challenging vicinity of metallic implants.
• The findings pave the way for more efficient and higher-quality MR imaging in patients with metal implants.