Based on the number of extremely large number registrations for the 1^st AI in an imaging core Town Hall and the breadth of this topic, we will be holding more Town Halls.  Since everyone attending felt the breakout sessions were too short, future Town Hall will focus on a single topic.

As someone who runs a lab or a core facility, have you ever wished you could clone yourself?  Have you ever itched for more time in your busy day to spend on deep-impact, brain-intensive tasks like grant and manuscript writing, helping people with better experimental design, etc.?  Do you feel like you are constantly being interrupted from your focus by answering basic questions for which your users and clients should already know the answers? 

During this session, I will describe how I have employed a chatbot to provide my users with a resource that is always available and has been trained to answer most of the basic questions I get.  It answers questions only from my own curated content, so there is no risk of it crawling the internet for random (wrong) information.  I will go over how I started down this path, the progress that I have made in improving the system, and where I see this project headed.

This session will be a combination of a brief presentation to provide the context needed, followed by a hands-on workshop with audience questions, and ending with ~10 minutes of discussion of collaborative ideas for expanding the system to community members.  Once you register, you will be provided with a few items of optional “homework” in case you want to explore some of the software before the workshop.  While this will be slightly light microscopy-centric, the concepts could be easily employed in any multi-user laboratory.

Future Town Hall topics

• 101 machine learning, AI and everything in between ( with real-life examples)
• Image analysis workflows—when and how to incorporate AI
• How to use specific tools, e.g. CellProfiler, Fiji/ImageJ, NIS Elements, what models do these tools currently support?
• How to validate an AI model (i.e., how do you know that your approach is “correct”)
• Architectures and models – advantages, disadvantages, applications, how to choose etc.