Intelligent Signal Analysis Using Case-Based Reasoning for Decision Support in Stress Management - Computational Intelligence in Healthcare: Advanced Methodologies - page 149

Chapter 8
Intelligent Signal Analysis Using Case-Based Reasoning
for Decision Support in Stress Management
Shahina Begum, Mobyen Uddin Ahmed, Peter Funk, and Ning Xiong
School of Innovation, Design and Engineering, Mälardalen University,
PO Box 883 SE-721 23, Västerås, Sweden
Tel.: [phone]; Fax: [phone]
[e-mail]
Abstract. Modern daily life triggers stress for many people in different every-
day situations. Consequently many people live with increased stress levels un-
der long durations. It is recognized that increased exposure to stress may cause
serious health problems if undiagnosed and untreated. One of the physiological
parameters for quantifying stress levels is finger temperature, which helps
clinicians with the diagnosis and treatment of stress. However, in practice, the
complex and varying nature of signals often makes it difficult and tedious for a
clinician and particularly less experienced clinicians to understand, interpret
and analyze complex and lengthy sequential measurements. There are very few
experts who are able to diagnose and predict stress-related problems; hence a
system that can help clinicians in diagnosing stress is important. In order to
provide decision support in stress management using this complex data source,
case-based reasoning (CBR) is applied as the main methodology to facilitate
experience reuse and decision justification. Feature extraction methods which
aim to yield compact representation of original signals into case indexes are in-
vestigated. A fuzzy technique is also incorporated into the system to perform
matching between the features derived from signals to better accommodate
vagueness and uncertainty inherently existing in clinicians reasoning as well as
imprecision in case description. The validation of the approach is based on
close collaboration with experts and measurements from twenty four people
used as a reference. The system formulates a new problem case with 17 ex-
tracted features from finger temperature signal data. Every case contains fifteen
minutes of data from 1800 samples. Thirty nine time series from 24 people have
been used to evaluate the approach (matching algorithms) in which the system
shows a level of performance close to an experienced expert. Consequently, the
system can be used as an expert for a less experienced clinician or as a second
option for an experienced clinician to supplement their decision making task in
stress management.
1 Introduction
Today's increased use of computer-based systems in the medical domain requires
computer processing of biomedical signals such as, ECG, EEG, EMG, heart rate etc.
These are often obtaining from sensors, transducers, etc. and such signals contain