AIR·MS Training
Fall 2025 Training Sessions
We will be holding 3 AIR·MS training sessions this Fall. These sessions will introduce you to the AIR·MS environment and AI-related tools through live demonstrations you can follow along with. Each session will be offered in a hybrid format, with our team onsite to provide support and answer questions. Material will be provided 1 week prior to each session for registered users.
Session 1: Getting Started with AIR·MS: Health Data Fundamentals
Tuesday September 30 (10:00 am – 11:00 am) | Hybrid In-Person & Zoom
RECORDING NOW AVAILABLE – Session 1: Getting Started with AIR·MS: Health Data Fundamentals
SLIDES NOW AVAILABLE HERE: (PART 1) (PART 2)
PLEASE NOTE: THE MATERIAL FOR AIR·MS TRAINING SESSION 1 IS HERE.
You will need to have prior access to Minerva and AIR·MS OMOP De-ID dataset to participate in certain areas of this training. If you don’t have yet, you can follow these steps (NB: to access SailPoint you will need to be connected to the Mount Sinai network either by being onsite or by using VPN):
- Check that you have an ISMMS (School Network) Account. If you don’t, you can request it on SailPoint by selecting “School Network Account”
- If you haven’t yet, request a Minerva Account filling out this form. Please note that if you are not a PI, your PI should provide approval via email
- Request access to AIR·MS OMOP De-ID dataset via SailPoint by selecting “AIR.MS Production MSDW OMOP De-ID (MSSM)”. A detailed how to guide can be found here.
More details of how to gain access can be found here: AIR‧MS: Getting Started | Scientific Computing and Data
- In person attendees: In person sessions will take place in Annenberg Classroom 10-70.
- Virtual attendees: Zoom links will be provided following registration for each session.
In session 1, participants will learn how to:
- Understand data flow from EPIC to AIR·MS.
- Avoid common pitfalls in working with clinical data.
- Perform exploratory data analysis (EDA) in AIR·MS.
Includes a live demo notebook shared in advance.
Audience: Researchers, data scientists, and clinicians new to AIR·MS.
Session 2: From ChatAI to AIR·MS: Leveraging Large Language Models
Tuesday October 7 (10:00 am – 11:00 am) | Hybrid In-Person & Zoom
RECORDING NOW AVAILABLE – Session 2: From ChatAI to AIR·MS: Leveraging Large Language Models
SLIDES NOW AVAILABLE HERE.
PLEASE NOTE: THE MATERIAL FOR AIR·MS TRAINING SESSION 2 IS HERE. The direct link to the Jupyter notebook is HERE.
REMINDER: Participants will need to request access to Minerva and AIR·MS by Friday October 3rd at the latest so that there is sufficient time to process applications. Participants will need to have a working SYMANTEC VIP 2-factor authentication.
- Zoom attendees: Register here for From ChatAI to AIR·MS: Leveraging Large Language Models (LLM)
- In-Person Attendees (Annenberg Classroom 10-68): Register here
Discover how on-premises open-source LLMs can support data exploration in AIR·MS. This session covers:
- ChatAI demo.
- Using Ollama and AIR·MS within Python.
- Introduction to SQL
- How LLMs can help write and refine SQL queries.
Audience: Introductory users interested in AI tools for health data.
Session 3: Advanced AIR·MS: Deep Dive into Data Modalities and AI/Machine Learning (ML) Applications
Tuesday October 21 (10:00 am – 11:00 am) | Hybrid In-Person & Zoom
- Zoom attendees: Register here for Advanced AIR·MS: Deep Dive into Data Modalities and AI/ML Applications
- In-Person Attendees (Annenberg Classroom 10-70): Register here
Explore advanced AIR·MS capabilities for multimodal research, with lightning talks on OMOP, pathology, radiology, echocardiography, and Mount Sinai Million.
- Includes an overview of ML infrastructure at Mount Sinai and a demo of multimodal AI workflows.
Audience: Advanced researchers and technical users.
Please register ahead of time for sessions. Direct any questions to airms-support@mssm.edu.
Fall 2025 Inaugural AIR‧MS Town Hall
The Fall 2025 Inaugural AIR‧MS Town Hall is now available (powerpoint slides) (video recording).
Spring 2025 Inaugural AIR‧MS Training Session
The Inaugural AIR‧MS Training video is now available here.