For the beginner track, micro-learning tutorials built on the UF AI Boot Camp curricula will be modified and taught by CHoRUS faculty and UF College of Medicine AI faculty. Trainees will be guided through several hands-on example applications of AI in Clinical Care via interactive Python programming notebooks that showcase AI/ML methods applied to real-world clinical datasets. Instructional material will combine didactic with experiential learning. The beginner track assumes no prior programming experience and is aimed primarily at AI/ML novices.
Workshop material will consist of brief case studies that articulate and evaluate AI/ML practices and contribute to the development of AI in Clinical Care domain knowledge. For this track, we will accept 40 total trainees. Thirty slots are open to practicing clinicians or students who are interested in obtaining a basic certification in AI in healthcare.
Before the workshop (self-paced prerequisite instruction will be sent to registrants in March):
- Introduction to Python (syntax, variables, libraries, loops, conditionals, functions)
- Introduction to Version Control: Reproducibility with git and Github
- FAIR Data for AI (Findable, Accessible, Interoperable, Reusable)
Schedule
Workshop Sessions will run in two blocks: 8:30 am-11:45 am and 1:30 pm-4:45 pm
- Review of AI Fundamentals Course and Kick-off
- Interactive sessions
- Python for Clinical Data Science (Scott Siegal and Yuanfang Ren)
- Developing a clinical decision support tool using machine learning and electronic health records (Ben Shickel and Yuanfang Ren)
- Special topics seminars with illustrated examples
- AI in Medical Imaging (Shao)
- AI in Multi-Omics (Graim)
- AI in Pathology (Sarder)
- AI with Clinical Text (Wu)
Learning Objectives
By the end of this workshop, students will be able to:
- Navigate the Jupyter Lab environment and create Jupyter notebooks
- Explain the rules that govern Python variables, loops, conditionals, and functions
- Develop and execute Python code for manipulating medical data
- Identify the important Python libraries for biomedical data science
- Explain the advantages of FAIR data and version control for AI collaboration
- Train a machine learning model to predict patient outcomes with structured medical datasets
- Describe the deep learning methods used for radiology, pathology, and text data
- Identify the challenges and opportunities for AI in medical imaging and NLP
- Discuss the importance of multidisciplinary collaboration for advancing medical AI
Participants: The Beginner Track can host 40 total participants:
- 10 seats: Recipients of CHoRUS members travel grants (details at https://chorus4ai.org/).
- 10 seats: Held for UF students, residents and fellows through a competitive process. These individuals will receive $500 to cover their travel and free registration.
- 20 seats: Open to all with a registration fee of $200.