step3.qmd

---
title: "Step 3: Annotate Slices"
editor: 
  markdown: 
    wrap: 72
---
Step 3: annotating is happeneing in parallel with Step 4: patching

**Aim: identify different classes in the epithelium so we can train a model with ground truth**

# Summary

Matched slices are sent to stakeholders for annotation with **Benign**, **Low Grade**, and **High Grade** C-MIL labels, providing ground truth for model training. This step runs **in parallel** with patching development.

We have sent matched slices to the stakeholder for them to annotate with Benign, Low grade, and High grade C-MIL. This will provide a ground truth for our future model. 

![](images/Annotations.png){fig-align="center"}

In the meantime, we are working on our patching algorithm.