Protocol to annotate and automate single-cell instance segmentation on stimulated Raman histology using deep learning.

Stimulated Raman histology (SRH) is a label-free optical imaging technique that can discern molecular components such as lipids and proteins at subcellular spatial resolution without histologic staining. Here, we present a protocol for labeling cells and training AI models for automated cell segmentation on SRH images acquired intra-operatively from neurosurgical cases. We describe steps to enable single-cell spatial analysis on SRH using ELUCIDATE, a web-based SRH cell annotation tool, and DetectSRH Python library.