Nutritional Diagnosis Presentation

Developing a Rule-based Expert System according to ADA’s Nutritional Care Process for Nutritional Diagnosis 台北醫學大學醫學資訊研究所 研究生：陳昱全 G158097008 指導教授：徐建業 老師 共同指導教授：劉　立 老師

Topics  Introduction • Literature Review  Proposed Research Methods and Design  Live Demo of the progress so far  Expected Difficulties  Expected Results  References

Introduction  Research Background • Gov’t Research project (DOH) • Assoc. Prof. Sherry Yang  Research Motivation • The importance of developing a national standard for dietetics in Taiwan. • To assist dietetics in making decisions for nutritional diagnosis. • To provide a tool for dietetics professionals for decision making while making a diagnosis.

Nutrition Diagnosis  Nutrition Diagnosis is the second step in the Nutritional Care Process (NCP) defined by the American Dietetics Association in Aug 2003.  Nutrition Diagnosis • Definition: A food and nutrition professional’s identification and labeling of an existing nutrition problem that the food and nutrition professional is responsible for treating independently.  NCP includes 4 separate but interconnected steps • Nutrition Assessment • Nutrition Diagnosis • Nutrition Intervention • Nutrition Monitoring and Evaluation (2008). "Nutrition Care Process and Model Part I: The 2008 Update.“ Journal of the American Dietetic Association 108(7): 1113-1117.

(2008). "Nutrition Care Process and Model Part I: The 2008 Update.“ Journal of the American Dietetic Association 108(7): 1113-1117.

Nutrition Diagnosis Outcomes  Nutrition Diagnosis PESS Statement • Problem ( Diagnostic Label ) • Etiology ( Cause/contributing factors ) • Signs/Symptoms ( Defining characteristics )  A well written PESS statement should be • Clear and concise • Specific to one patient • Limited to one problem • Based on signs and symptoms from assessment data

(2008). "Nutrition Care Process and Model Part I: The 2008 Update.“ Journal of the American Dietetic Association 108(7): 1113-1117.

Nutrition Diagnosis Outcomes

(2008). "Nutrition Care Process and Model Part I: The 2008 Update.“ Journal of the American Dietetic Association 108(7): 1113-1117.

Expert System  A computer program that provides expert-level solutions to important problems. • Heuristic • Transparent • flexible  Attempt to imitate expert reasoning processes and knowledge in solving specific problem.  Most popular applied AI technology • Enhance productivity • Augment work forces  Expert System do not replace Experts, but they • Make their knowledge and experience more widely available • Permit non-experts to work better

(2008). "Nutrition Care Process and Model Part I: The 2008 Update.“ Journal of the American Dietetic Association 108(7): 1113-1117.

Expert System - Expertise  The extensive, task specific knowledge acquired from training, reading, and experience • Theories • Procedures (Rules) • Facts  Enable experts to be better and faster than nonexperts  Expertise is usually associated with vast quantity of knowledge.  Expert knowledge is well stored, organized and provides a way for easy retrieval from an expert.  Experts have excellent memory.

(2008). "Nutrition Care Process and Model Part I: The 2008 Update.“ Journal of the American Dietetic Association 108(7): 1113-1117.

Human Experts        

Recognize and formulate the problem Solve problems quickly and properly Explain the solution Learn from experience Restructure knowledge Break rules Determine relevance Degrade

(2008). "Nutrition Care Process and Model Part I: The 2008 Update.“ Journal of the American Dietetic Association 108(7): 1113-1117.

Transferring Expertise  Objective of an expert system • To transfer expertise from experts to a computer system  Activities • Knowledge acquisition • Knowledge representation • Knowledge inferencing (Reasoning/Thinking ) • Knowledge transfer to the user  Knowledge is stored in a knowledge base • Facts • Rules( Procedures )

(2008). "Nutrition Care Process and Model Part I: The 2008 Update.“ Journal of the American Dietetic Association 108(7): 1113-1117.

Two knowledge Types  Facts • Format: The of