Feature selection on chronic pain self reporting data

Yan Huang, Huiru Zheng, Christopher Nugent, Paul McCullagh, Norman Black, Kevin Vowles, Lance McCracken

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

Chronic pain is a common long-term condition that changes patients' physical and emotional functioning. Currently, the integrated biopsychosoical approach is the mainstay treatment for patients with chronic pain. Self reporting (the use of questionnaires) is one of the most common methods to evaluate treatment outcome. Nevertheless, a large number of questions (for example 329 questions in this study) may be required and as such may be viewed as not being convenient for patients to complete. This paper has applied the theory of information gain to rank the questions in addition to investigating important factors related to the treatment outcome. Analysis within the study ranked the questions from 1 to 329 based on information gain (highest to lowest). Results showed that questions related to chronic pain coping strategies and value-based actions had high information gain. Four supervised learning classifiers were used to investigate the correlations between feature numbers and classification accuracy. The results showed classifier that a multi-layer perceptron classifier obtained the highest classification accuracy (96.05%) on an optimized subset which consisted of 133 questions.
LanguageEnglish
Title of host publicationUnknown Host Publication
Pages1-4
Number of pages4
DOIs
Publication statusPublished - 2009
EventInformation Technology and Applications in Biomedicine, 2009. ITAB 2009. 9th International Conference - Cyprus
Duration: 1 Jan 2009 → …

Conference

ConferenceInformation Technology and Applications in Biomedicine, 2009. ITAB 2009. 9th International Conference
Period1/01/09 → …

Fingerprint

Chronic Pain
Research Design
Information Theory
Neural Networks (Computer)
Learning
Therapeutics
Surveys and Questionnaires

Cite this

Huang, Y., Zheng, H., Nugent, C., McCullagh, P., Black, N., Vowles, K., & McCracken, L. (2009). Feature selection on chronic pain self reporting data. In Unknown Host Publication (pp. 1-4) https://doi.org/10.1109/ITAB.2009.5394419
Huang, Yan ; Zheng, Huiru ; Nugent, Christopher ; McCullagh, Paul ; Black, Norman ; Vowles, Kevin ; McCracken, Lance. / Feature selection on chronic pain self reporting data. Unknown Host Publication. 2009. pp. 1-4
@inproceedings{079acecf7d3d45e0914ac5d091292b78,
title = "Feature selection on chronic pain self reporting data",
abstract = "Chronic pain is a common long-term condition that changes patients' physical and emotional functioning. Currently, the integrated biopsychosoical approach is the mainstay treatment for patients with chronic pain. Self reporting (the use of questionnaires) is one of the most common methods to evaluate treatment outcome. Nevertheless, a large number of questions (for example 329 questions in this study) may be required and as such may be viewed as not being convenient for patients to complete. This paper has applied the theory of information gain to rank the questions in addition to investigating important factors related to the treatment outcome. Analysis within the study ranked the questions from 1 to 329 based on information gain (highest to lowest). Results showed that questions related to chronic pain coping strategies and value-based actions had high information gain. Four supervised learning classifiers were used to investigate the correlations between feature numbers and classification accuracy. The results showed classifier that a multi-layer perceptron classifier obtained the highest classification accuracy (96.05{\%}) on an optimized subset which consisted of 133 questions.",
author = "Yan Huang and Huiru Zheng and Christopher Nugent and Paul McCullagh and Norman Black and Kevin Vowles and Lance McCracken",
year = "2009",
doi = "10.1109/ITAB.2009.5394419",
language = "English",
isbn = "978-1-4244-5379-5",
pages = "1--4",
booktitle = "Unknown Host Publication",

}

Huang, Y, Zheng, H, Nugent, C, McCullagh, P, Black, N, Vowles, K & McCracken, L 2009, Feature selection on chronic pain self reporting data. in Unknown Host Publication. pp. 1-4, Information Technology and Applications in Biomedicine, 2009. ITAB 2009. 9th International Conference, 1/01/09. https://doi.org/10.1109/ITAB.2009.5394419

Feature selection on chronic pain self reporting data. / Huang, Yan; Zheng, Huiru; Nugent, Christopher; McCullagh, Paul; Black, Norman; Vowles, Kevin; McCracken, Lance.

Unknown Host Publication. 2009. p. 1-4.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Feature selection on chronic pain self reporting data

AU - Huang, Yan

AU - Zheng, Huiru

AU - Nugent, Christopher

AU - McCullagh, Paul

AU - Black, Norman

AU - Vowles, Kevin

AU - McCracken, Lance

PY - 2009

Y1 - 2009

N2 - Chronic pain is a common long-term condition that changes patients' physical and emotional functioning. Currently, the integrated biopsychosoical approach is the mainstay treatment for patients with chronic pain. Self reporting (the use of questionnaires) is one of the most common methods to evaluate treatment outcome. Nevertheless, a large number of questions (for example 329 questions in this study) may be required and as such may be viewed as not being convenient for patients to complete. This paper has applied the theory of information gain to rank the questions in addition to investigating important factors related to the treatment outcome. Analysis within the study ranked the questions from 1 to 329 based on information gain (highest to lowest). Results showed that questions related to chronic pain coping strategies and value-based actions had high information gain. Four supervised learning classifiers were used to investigate the correlations between feature numbers and classification accuracy. The results showed classifier that a multi-layer perceptron classifier obtained the highest classification accuracy (96.05%) on an optimized subset which consisted of 133 questions.

AB - Chronic pain is a common long-term condition that changes patients' physical and emotional functioning. Currently, the integrated biopsychosoical approach is the mainstay treatment for patients with chronic pain. Self reporting (the use of questionnaires) is one of the most common methods to evaluate treatment outcome. Nevertheless, a large number of questions (for example 329 questions in this study) may be required and as such may be viewed as not being convenient for patients to complete. This paper has applied the theory of information gain to rank the questions in addition to investigating important factors related to the treatment outcome. Analysis within the study ranked the questions from 1 to 329 based on information gain (highest to lowest). Results showed that questions related to chronic pain coping strategies and value-based actions had high information gain. Four supervised learning classifiers were used to investigate the correlations between feature numbers and classification accuracy. The results showed classifier that a multi-layer perceptron classifier obtained the highest classification accuracy (96.05%) on an optimized subset which consisted of 133 questions.

U2 - 10.1109/ITAB.2009.5394419

DO - 10.1109/ITAB.2009.5394419

M3 - Conference contribution

SN - 978-1-4244-5379-5

SP - 1

EP - 4

BT - Unknown Host Publication

ER -