SpatialFusion

This method is described in the paper (TBD).

SpatialFusion is a lightweight foundation model designed to represent find niches in tissue using a lower dimensional embedding. It integrates spatial transcriptomics data with histopathology-derived image features into a shared latent representation, and can be applied to paired spatial transcriptomics and whole slide images or whole slide images only.

This documentation covers: - Installation instructions - A quick start guide - Conceptual background - Tutorials and examples - API reference

Source code: https://github.com/uhlerlab/spatialfusion