A Machine Learning Approach for the Identification of a Biomarker of Human Pain using fNIRS

Raul Fernandez Rojas, Xu Huang, Keng Liang Ou

Research output: Contribution to journalArticle

2 Citations (Scopus)
4 Downloads (Pure)

Abstract

Pain is a highly unpleasant sensory and emotional experience, and no objective diagnosis test exists to assess it. In clinical practice there are two main methods for the estimation of pain, a patient’s self-report and clinical judgement. However, these methods are highly subjective and the need of biomarkers to measure pain is important to improve pain management, reduce risk factors, and contribute to a more objective, valid, and reliable diagnosis. Therefore, in this study we propose the use of functional near-infrared spectroscopy (fNIRS) and machine learning for the identification of a possible biomarker of pain. We collected pain information from 18 volunteers using the thermal test of the quantitative sensory testing (QST) protocol, according to temperature level (cold and hot) and pain intensity (low and high). Feature extraction was completed in three different domains (time, frequency, and wavelet), and a total of 69 features were obtained. Feature selection was carried out according to three criteria, information gain (IG), joint mutual information (JMI), and Chi-squared (χ 2 ). The significance of each feature ranking was evaluated using three learning models separately, linear discriminant analysis (LDA), the K-nearest neighbour (K-NN) and support vector machines (SVM) using the linear and Gaussian and polynomial kernels. The results showed that the Gaussian SVM presented the highest accuracy (94.17%) using only 25 features to identify the four types of pain in our database. In addition, we propose the use of the top 13 features according to the JMI criteria, which exhibited an accuracy of 89.44%, as promising biomarker of pain. This study contributes to the idea of developing an objective assessment of pain and proposes a potential biomarker of human pain using fNIRS.

Original languageEnglish
Article number5645
Pages (from-to)1-12
Number of pages12
JournalScientific Reports
Volume9
DOIs
Publication statusPublished - 4 Apr 2019

Fingerprint

Near-Infrared Spectroscopy
Biomarkers
Pain
Hot Temperature
Joints
Machine Learning
Discriminant Analysis
Pain Measurement
Pain Management
Self Report
Volunteers
Linear Models
Learning
Databases

Cite this

Fernandez Rojas, Raul ; Huang, Xu ; Ou, Keng Liang. / A Machine Learning Approach for the Identification of a Biomarker of Human Pain using fNIRS. In: Scientific Reports. 2019 ; Vol. 9. pp. 1-12.
@article{44a85f7ca08945c1ad0674e4c596d194,
title = "A Machine Learning Approach for the Identification of a Biomarker of Human Pain using fNIRS",
abstract = "Pain is a highly unpleasant sensory and emotional experience, and no objective diagnosis test exists to assess it. In clinical practice there are two main methods for the estimation of pain, a patient’s self-report and clinical judgement. However, these methods are highly subjective and the need of biomarkers to measure pain is important to improve pain management, reduce risk factors, and contribute to a more objective, valid, and reliable diagnosis. Therefore, in this study we propose the use of functional near-infrared spectroscopy (fNIRS) and machine learning for the identification of a possible biomarker of pain. We collected pain information from 18 volunteers using the thermal test of the quantitative sensory testing (QST) protocol, according to temperature level (cold and hot) and pain intensity (low and high). Feature extraction was completed in three different domains (time, frequency, and wavelet), and a total of 69 features were obtained. Feature selection was carried out according to three criteria, information gain (IG), joint mutual information (JMI), and Chi-squared (χ 2 ). The significance of each feature ranking was evaluated using three learning models separately, linear discriminant analysis (LDA), the K-nearest neighbour (K-NN) and support vector machines (SVM) using the linear and Gaussian and polynomial kernels. The results showed that the Gaussian SVM presented the highest accuracy (94.17{\%}) using only 25 features to identify the four types of pain in our database. In addition, we propose the use of the top 13 features according to the JMI criteria, which exhibited an accuracy of 89.44{\%}, as promising biomarker of pain. This study contributes to the idea of developing an objective assessment of pain and proposes a potential biomarker of human pain using fNIRS.",
keywords = "Biomarker, Pain, fNIRS",
author = "{Fernandez Rojas}, Raul and Xu Huang and Ou, {Keng Liang}",
year = "2019",
month = "4",
day = "4",
doi = "10.1038/s41598-019-42098-w",
language = "English",
volume = "9",
pages = "1--12",
journal = "Scientific Reports",
issn = "2045-2322",
publisher = "Nature Publishing Group",

}

A Machine Learning Approach for the Identification of a Biomarker of Human Pain using fNIRS. / Fernandez Rojas, Raul; Huang, Xu; Ou, Keng Liang.

In: Scientific Reports, Vol. 9, 5645, 04.04.2019, p. 1-12.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A Machine Learning Approach for the Identification of a Biomarker of Human Pain using fNIRS

AU - Fernandez Rojas, Raul

AU - Huang, Xu

AU - Ou, Keng Liang

PY - 2019/4/4

Y1 - 2019/4/4

N2 - Pain is a highly unpleasant sensory and emotional experience, and no objective diagnosis test exists to assess it. In clinical practice there are two main methods for the estimation of pain, a patient’s self-report and clinical judgement. However, these methods are highly subjective and the need of biomarkers to measure pain is important to improve pain management, reduce risk factors, and contribute to a more objective, valid, and reliable diagnosis. Therefore, in this study we propose the use of functional near-infrared spectroscopy (fNIRS) and machine learning for the identification of a possible biomarker of pain. We collected pain information from 18 volunteers using the thermal test of the quantitative sensory testing (QST) protocol, according to temperature level (cold and hot) and pain intensity (low and high). Feature extraction was completed in three different domains (time, frequency, and wavelet), and a total of 69 features were obtained. Feature selection was carried out according to three criteria, information gain (IG), joint mutual information (JMI), and Chi-squared (χ 2 ). The significance of each feature ranking was evaluated using three learning models separately, linear discriminant analysis (LDA), the K-nearest neighbour (K-NN) and support vector machines (SVM) using the linear and Gaussian and polynomial kernels. The results showed that the Gaussian SVM presented the highest accuracy (94.17%) using only 25 features to identify the four types of pain in our database. In addition, we propose the use of the top 13 features according to the JMI criteria, which exhibited an accuracy of 89.44%, as promising biomarker of pain. This study contributes to the idea of developing an objective assessment of pain and proposes a potential biomarker of human pain using fNIRS.

AB - Pain is a highly unpleasant sensory and emotional experience, and no objective diagnosis test exists to assess it. In clinical practice there are two main methods for the estimation of pain, a patient’s self-report and clinical judgement. However, these methods are highly subjective and the need of biomarkers to measure pain is important to improve pain management, reduce risk factors, and contribute to a more objective, valid, and reliable diagnosis. Therefore, in this study we propose the use of functional near-infrared spectroscopy (fNIRS) and machine learning for the identification of a possible biomarker of pain. We collected pain information from 18 volunteers using the thermal test of the quantitative sensory testing (QST) protocol, according to temperature level (cold and hot) and pain intensity (low and high). Feature extraction was completed in three different domains (time, frequency, and wavelet), and a total of 69 features were obtained. Feature selection was carried out according to three criteria, information gain (IG), joint mutual information (JMI), and Chi-squared (χ 2 ). The significance of each feature ranking was evaluated using three learning models separately, linear discriminant analysis (LDA), the K-nearest neighbour (K-NN) and support vector machines (SVM) using the linear and Gaussian and polynomial kernels. The results showed that the Gaussian SVM presented the highest accuracy (94.17%) using only 25 features to identify the four types of pain in our database. In addition, we propose the use of the top 13 features according to the JMI criteria, which exhibited an accuracy of 89.44%, as promising biomarker of pain. This study contributes to the idea of developing an objective assessment of pain and proposes a potential biomarker of human pain using fNIRS.

KW - Biomarker

KW - Pain

KW - fNIRS

UR - http://www.scopus.com/inward/record.url?scp=85063987734&partnerID=8YFLogxK

UR - http://www.mendeley.com/research/machine-learning-approach-identification-biomarker-human-pain-using-fnirs

U2 - 10.1038/s41598-019-42098-w

DO - 10.1038/s41598-019-42098-w

M3 - Article

VL - 9

SP - 1

EP - 12

JO - Scientific Reports

JF - Scientific Reports

SN - 2045-2322

M1 - 5645

ER -