Clustering according to image features

[gboscagli]

Greetings,
first of all, thanks for developing the Squidpy tool.
I'm new to this tool and I was wondering if there is a way to "merge" the information coming from gene expression and image features to drive the clustering. As per now, I saw that the two types of clustering run separately and are then compared. Is it something that can be done using Squidpy, or perhaps is in the development roadmap? Is it even a reasonable and/or reliable thing to perform, given the heterogeneity of the information coming from gene expression and graphical features?