# Pathology Foundation Models

Frozen pathology foundation-model embeddings transfer poorly to single-cell spatial gene expression prediction from H&E: on the 5K gene panel, mean per-gene Pearson correlation falls from 0.375 at pseudo-spot resolution to 0.065 at single-cell resolution. A set of single-cell-oriented lightweight adaptations, namely cell-localized token pooling, 40x field-of-view inputs, Poisson count supervision and LoRA fine-tuning, raises the same panel to 0.174 and beats SpatialEx on the 5K panel with AUROC 0.800 versus 0.642. Gains concentrate in spatially variable and disease-relevant genes, while absolute correlation across the full panel stays low.

Frozen pathology foundation-model embeddings transfer poorly to single-cell spatial gene expression prediction from H&E: on the 5K gene panel, mean per-gene Pearson correlation falls from 0.375 at pseudo-spot resolution to 0.065 at single-cell resolution. A set of single-cell-oriented lightweight adaptations, namely cell-localized token pooling, 40x field-of-view inputs, Poisson count supervision and LoRA fine-tuning, raises the same panel to 0.174 and beats SpatialEx on the 5K panel with AUROC 0.800 versus 0.642. Gains concentrate in spatially variable and disease-relevant genes, while absolute correlation across the full panel stays low.

The authors introduce SlideRuler, which treats other regions of the same slide as internal controls and learns a transfer map from paired rescans to estimate and correct acquisition-induced shifts, enabling calibration from a single scan at inference while keeping the foundation model frozen; across two encoders and five SCORPION scanners, learned transfer reduces mean target-to-source embedding distance by 16.3–38.5% relative to raw embeddings, same-slide controls outperform unrelated same-scanner controls, and a source-anchored variant reduces source-feature displacement by 47.7–83.6% while retaining most of its alignment gain.

The authors introduce SlideRuler, which treats other regions of the same slide as internal controls and learns a transfer map from paired rescans to estimate and correct acquisition-induced shifts, enabling calibration from a single scan at inference while keeping the foundation model frozen; across two encoders and five SCORPION scanners, learned transfer reduces mean target-to-source embedding distance by 16.3–38.5% relative to raw embeddings, same-slide controls outperform unrelated same-scanner controls, and a source-anchored variant reduces source-feature displacement by 47.7–83.6% while retaining most of its alignment gain.