Neural field patient-specific super resolution for enhanced 1.5 Tesla brain MRI visualization

Brain magnetic resonance imaging serves as a cornerstone of preoperative neurosurgical assessment. Neural fields represent an emerging machine learning approach capable of super-resolution reconstruction and novel view synthesis without requiring large training datasets. We evaluated ten 1.5-Tesla brain MRI sequences (nine anisotropic and one isotropic) to train patient-specific neural field models using a proprietary framework (Radscaler©). Image quality assessment was performed on reconstructions upscaled by factors of 2×, 3×, and 4× relative to original resolution. The method achieved favorable quality metrics across all scaling factors: mean SSIM of 0.85 (±0.04), MS-SSIM of 0.95 (±0.01), and LPIPS of 0.09 (±0.04). Neural field reconstruction enabled enhanced visualization of micro-anatomical structures through improved spatial resolution and interpolation of intermediate views not present in the original acquisition. These findings demonstrate that neural fields provide a clinically viable approach for volumetric MRI super-resolution and novel view synthesis, particularly valuable for addressing anisotropic acquisition limitations in neurosurgical planning.