Year: 2010
Type: article
Abstract: Adaptation is a difficult part of the case-based reasoning (CBR) cycle that is often omitted in medical domains.The CBR paradigm solves new problems based on the solutions to similar, past problems.Adaptation is the process of modifying a past solution to fit a new problem.This thesis identifies the components of solutions and the ways in which they can be modified for an intelligent decision support system for type 1 diabetes management.Type 1 diabetes is an autoimmune disease which affects nearly one million Americans.Maintaining good blood glucose control is essential for these patients to avoid serious complications of the disease.A prototypical CBR system has been built that identifies problems the patient is experiencing with blood glucose control and finds similar past problems with solutions to alleviate these problems.This research was conducted in the context of this prototypical system.This thesis contributes the design, implementation and evaluation of an adaptation module, which tailors past advice to a specific patient for a particular problem.Previously, only a closely matching problem and its unmodified solution could be viewed.Physicians verified that adapted solutions were better than unmodified solutions 83% of the time.They also identified the types of adaptation that were most useful to physicians and patients.A secondary contribution is a new graphical user interface, which presents the adapted solutions along with other tools to aid in patient care.I would like to give a huge thank you to all the people who have supported me through my education and the writing of this thesis.I sincerely appreciate the love, support and encouragement my friends and family have given me.I can't put into words how much their faith in me has meant while doing this research.I would also like to thank all those responsible for my education.It is because of them that I was able to get to this point.
Author: Tessa Cooper
Institution: Ohio University
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W13037908