30
Sep 2024
AI Passport for Biomedical Research
Agentic AI
The landscape of artificial intelligence is rapidly evolving, moving beyond traditional models to embrace a new paradigm: agentic AI. These aren’t simply systems that respond to instructions; they’re capable of autonomous action, independent planning, and proactive goal pursuit. This shift presents a transformative opportunity across countless fields, and we’re excited to introduce a program designed to help you harness its power. Many researchers, however, are just beginning to explore this frontier. Traditional AI training often focuses on passive systems, leaving a gap in understanding how to effectively build with, and leverage, these more dynamic, agentic approaches. Our program directly addresses this need, providing the essential skills to navigate this exciting new territory and unlock the potential of autonomous AI agents within your own research and development efforts. We believe this is a foundational step towards a future of accelerated discovery and innovation – and we’re thrilled to help you be a part of it.
Course Outcomes
Integration of AI Techniques:
Participants will be able to integrate AI and machine learning techniques into their healthcare research and clinical practice.
Application of Real-World Data:
Learners will effectively use real-world medical data and case studies to address healthcare challenges and develop innovative solutions.
Practical AI Skills:
Graduates will gain hands-on experience with code-free AI/ML tools, enabling them to implement AI solutions without extensive programming knowledge.
Critical Thinking and Problem Solving:
Participants will enhance their ability to analyze and solve complex healthcare problems through AI-driven insights and methodologies.
Module 1: Fundamentals of Biomedical AI Research
Understand the capabilities, limitations, and applications of biomedical AI and AI's lifecycle
Microskill:
- Demystifying artificial intelligence
- Artificial intelligence lifecycle
- Designing biomedical artificial intelligence experiments
- Training, validation, and generalizability
- Leveraging multidisciplinary team strengths
- Basics of scientific rigor and reproducibility
- Mentorship and peer review in biomedical AI
Module 2: Biomedical AI Alignment
Explore the ethics of biomedical AI, including bias, fairness, liability, and societal impact
Microskill :
- Fundamentals of biomedical AI ethics and liability
- Bias, fairness, and societal impact of biomedical AI
- The regulatory landscape of biomedical AI
- Biomedical AI quality and safety
- Human-AI collaboration in biomedicine
- Human-AI collaboration in biomedicine
- The fundamental principles of bioethics
Module 3: Data-Centric Biomedical AI
Explore the principles of developing an AI-ready biomedical dataset
Microskill:
- The importance of data for developing biomedical AI
- Acquiring ethically sourced biomedical data
- Understanding the role of human annotation
- Promoting FAIR biomedical data principles
- Developing AI/ML-ready biomedical datasets
- Navigating multi-institutional data sharing challenges
- Secure and ethical use of biomedical data
Module 4: Fundamentals of Biomedical ML
Evaluate the basics of machine learning and how to choose the right ML/deep learning model
Microskill:
- Shared biomedical artificial intelligence vocabulary
- Applied fundamentals of ML and deep learning
- Choosing the right biomedical machine learning model
- Choosing the right biomedical deep learning model
- Evaluating biomedical machine learning models
- Model generalizability
- Ethics of black-box algorithms
Module 5: Biomedical Image Analysis
Learn about the process and applications of biomedical AI image analysis
Microskill:
- Landscape of biomedical imaging
- Biomedical image preprocessing and transformation
- Traditional biomedical image analysis
- Biomedical computer vision applications
- Advanced and emerging topics
- Consistency in biomedical image analysis
- Ethical and privacy implications of biomedical imaging
Module 6: Generative AI in Biomedicine
Understand the fundamentals of generative biomedical AI and large language models (LLMs)
Microskill:
- Fundamentals of generative biomedical AI
- Fundamentals of large language models
- Large language models (LLMs) in biomedicine
- Prompt engineering for biomedical applications
- Utilizing LLMs for accelerating biomedical research
- Evaluation and reproducibility of AI-generated data
- Ethical dissemination of generated biomedical content
Module 7: Impact Project
Learn to design biomedical AI experiments, write successful proposals, effectively communicate research, and incorporate biomedical AI into traditional research
Microskill:
- Designing biomedical AI experiments
- Writing successful biomedical AI proposals
- Effective scientific communication
- Bridging traditional research with AI innovation
- Peer review and feedback mechanisms
- Robust biomedical AI research design
- Responsible biomedical AI research
Module 8: Agentic AI
Microskill:
- Designing biomedical AI experiments
- Writing successful biomedical AI proposals
- Effective scientific communication
- Bridging traditional research with AI innovation
- Peer review and feedback mechanisms
- Robust biomedical AI research design
- Responsible biomedical AI research
Meet Your Instructors
Senior Associate Dean for Research Affairs
R. Glenn Davis Professor of Medicine,
Surgery and Anesthesiology Director,
Intelligent Clinical Care Center
Assistant Professor
Associate Director, Intelligent Clinical
Care Center
Associate Professor of Surgery
Associate Director, Intelligent Clinical
Care Center
Instructional Associate Professor
Elizabeth Palmer PhD
Assistant Director of Training and Education
General Surgery Resident
Yingbo Ma PhD
Data Scientist
Assistant Professor of Anesthesiology,
Clinical Director of research for the
Division of Critical Care Medicine