I wrote up a tutorial for how you can create a custom labeling and annotation platform for HCS data with bounding boxes and/or semantic segmentation. The annotation files are JSON files which can be manipulated during a CellProfiler pipeline or other ML process. I demonstrated this with C Elegans data, but you can use any dataset.
If you prefer to watch rather than read I created a video demonstration - https://youtu.be/KxpOKUSVwjk
And thanks to the Broad BioImage Benchmark Collection for supplying the dataset!
If you have any questions please don’t hesitate to reach out to me at email@example.com.