Introduction: Surgical pathology relies on destructive sectioning and delayed review of two-dimensional slides, limiting rapid whole-specimen assessment of spatially heterogeneous tissue. Fast three-dimensional imaging may enable pre-slide tissue characterization and more efficient workflows.

Methods: We performed high-resolution three-dimensional magnetic resonance histology (MRH) at 40 µm isotropic resolution in formalin-fixed surgical prostate cancer specimens (n=40). Digitized H&E slides were registered to multiparametric MRH (Figure 1) for spatially matched pathomic analysis. Quantitative cellular and tissue architecture features were compared with 203 MRH contrasts using linear and multivariable modeling. A pilot nnUNet model was trained to segment tumor, benign parenchyma, and stroma directly from MRH.

Results: Whole-specimen MRH was achieved in as little as 1.25 hours without exogenous dyes, tissue clearing, or specialized reagents. Multiple MRH contrasts demonstrated meaningful relationships with individual histologic features, with strongest single-feature associations reaching R²>0.6 across specimens (Figure 2). Multivariable models further improved performance (maximum R²=0.67), indicating that MRH signal reflects combined tissue properties rather than single metrics alone. Diffusion-sensitive contrasts were most strongly associated with cell density and spatial topology, and feature importance shifted with increasing b-values. Radiomic texture features extracted from MRH also demonstrated distinct histologic correlates. Initial three-dimensional tissue compartment segmentation demonstrated early feasibility (Dice=0.50) (Figure 3).

Conclusions: Fast, high-resolution MRH provides biologically interpretable volumetric tissue information from intact surgical specimens. This technology can support targeted sectioning, faster specimen triage, automated pre-slide analysis, and ultimately more efficient case review.

Learning Objectives:

1. Describe the basic principles of fast, 3D tissue imaging with magnetic resonance histology (MRH).
2. Draw parallels between measurable histology features and MRH signals.
3. Describe opportunities for improving pre-slide tissue analysis without workflow disruptions using MRH in surgical pathology.