BMI Course Information
BMI 311 Modeling Biomedical Knowledge (3)
Instructor Information |
|
Fall 2009 |
Course not offered |
Spring 2009 |
Course not offered |
Catalog Description
This course introduces concepts of artificial intelligence and knowledge modeling using medical informatics examples. Students will understand the historical foundations and motivations of AI in medical applications, learn how problem-solving, reasoning, knowledge management, and planning can be applied to medical informatics problems.
Prerequisites
CSE 100 or 110 and BMI 211
Textbook and Other Materials
Russell, S. J., Norvig, P., Canny, J. F., Malik, J., & Edwards, D. D.
(1995). Artificial intelligence: a modern approach. Prentice Hall Englewood Cliffs, NJ.
Course Learning Outcomes
Students who successfully complete the course will:
1. Understand and apply different search strategies to solve medical informatics problems.
2. Describe different approaches to reasoning.
3. Use logical, heuristic probabilistic and semantic methods of knowledge base construction.
4. Demonstrate different agent-based approaches to planning and AI in medical applications.
Major Topics and Time Covered
- Artificial Intelligence (1 week)
What is AI?
The foundations and history of AI in biomedical informatics (MYCIN, other expert systems)
- Problem solving by search (4 weeks)
Formulating problems
Search strategies
Informed search methods
Clinical diagnosis as a search problem
- Knowledge and reasoning (3 weeks)
Logical reasoning and first-order logic
Building a knowledge base
Inference
MYCIN's inference engine
Inference rules in hospital systems
- Logical Reasoning systems (3 weeks)
Prolog, theorem provers, production systems
Frame and semantic networks
The UMLS semantic network
Reasoning under uncertainty
Modeling uncertainty in clinical problems
- Planning systems (3 weeks)
Representations for states, goals and actions
Resource constraints
Planning and acting (conditional planning, plan execution)