Photoplethysmography temporal marker-based machine learning classifier for anesthesia drug detection

Syed Ghufran Khalid; Syed Mehmood Ali; Haipeng Liu; Aisha Ghazal Qurashi; Uzma Ali

doi:10.1007/s11517-022-02658-1

Photoplethysmography temporal marker-based machine learning classifier for anesthesia drug detection

Syed Ghufran Khalid, Syed Mehmood Ali, Haipeng Liu, Aisha Ghazal Qurashi, Uzma Ali

Research output: Contribution to journal › Article › peer-review

8 Citations (Scopus)

13 Downloads (Pure)

Abstract

Anesthesia drug overdose hazards and lack of gold standards in anesthesia monitoring lead to an urgent need for accurate anesthesia drug detection. To investigate the PPG waveform features affected by anesthesia drugs and develop a machine-learning classifier with high anesthesia drug sensitivity. This study used 64 anesthesia and non-anesthesia patient data (32 cases each), extracted from Queensland and MIMIC-II databases, respectively. The key waveform features (total area, rising time, width 75%, 50%, and 25%) were extracted from 16,310 signal recordings (5-s duration). Discriminant analysis, support vector machine (SVM), and K-nearest neighbor (KNN) were evaluated by splitting the dataset into halve training (11 patients, 8570 segments) and halve testing dataset (11 patients, 7740 segments). Significant differences exist between PPG waveform features of anesthesia and non-anesthesia groups (p 0.05). The KNN classifier achieved 91.7% (AUC = 0.95) anesthesia detection accuracy with the highest sensitivity (0.88) and specificity (0.90) as compared to other classifiers. Kohen's kappa also shows almost perfect agreement (0.79) with the KNN classifier. The KNN classifier trained with significant PPG features has the potential to be used as a reliable, non-invasive, and low-cost method for the detection of anesthesia drugs for depth analysis during surgical operations and postoperative monitoring. [Abstract copyright: © 2022. The Author(s).]

Original language	English
Pages (from-to)	3057-3068
Number of pages	12
Journal	Medical & biological engineering & computing
Volume	60
Issue number	11
Early online date	5 Sept 2022
DOIs	https://doi.org/10.1007/s11517-022-02658-1
Publication status	Published - Nov 2022

Bibliographical note

This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated
otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

Keywords

Queensland database
MIMIC II database
Anesthesia depth
K-nearest neighbor
Photoplethysmography

Access to Document

10.1007/s11517-022-02658-1

PublishedFinal published version, 1.89 MBLicence: CC BY

Cite this

@article{2e872bef499d4d1e836b1b5b01a2415d,

title = "Photoplethysmography temporal marker-based machine learning classifier for anesthesia drug detection",

abstract = "Anesthesia drug overdose hazards and lack of gold standards in anesthesia monitoring lead to an urgent need for accurate anesthesia drug detection. To investigate the PPG waveform features affected by anesthesia drugs and develop a machine-learning classifier with high anesthesia drug sensitivity. This study used 64 anesthesia and non-anesthesia patient data (32 cases each), extracted from Queensland and MIMIC-II databases, respectively. The key waveform features (total area, rising time, width 75%, 50%, and 25%) were extracted from 16,310 signal recordings (5-s duration). Discriminant analysis, support vector machine (SVM), and K-nearest neighbor (KNN) were evaluated by splitting the dataset into halve training (11 patients, 8570 segments) and halve testing dataset (11 patients, 7740 segments). Significant differences exist between PPG waveform features of anesthesia and non-anesthesia groups (p 0.05). The KNN classifier achieved 91.7% (AUC = 0.95) anesthesia detection accuracy with the highest sensitivity (0.88) and specificity (0.90) as compared to other classifiers. Kohen's kappa also shows almost perfect agreement (0.79) with the KNN classifier. The KNN classifier trained with significant PPG features has the potential to be used as a reliable, non-invasive, and low-cost method for the detection of anesthesia drugs for depth analysis during surgical operations and postoperative monitoring. [Abstract copyright: {\textcopyright} 2022. The Author(s).]",

keywords = "Queensland database, MIMIC II database, Anesthesia depth, K-nearest neighbor, Photoplethysmography",

author = "Khalid, {Syed Ghufran} and Ali, {Syed Mehmood} and Haipeng Liu and Qurashi, {Aisha Ghazal} and Uzma Ali",

note = "This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.",

year = "2022",

month = nov,

doi = "10.1007/s11517-022-02658-1",

language = "English",

volume = "60",

pages = "3057--3068",

journal = "Medical & biological engineering & computing",

issn = "0140-0118",

publisher = "Springer Verlag",

number = "11",

}

TY - JOUR

T1 - Photoplethysmography temporal marker-based machine learning classifier for anesthesia drug detection

AU - Khalid, Syed Ghufran

AU - Ali, Syed Mehmood

AU - Liu, Haipeng

AU - Qurashi, Aisha Ghazal

AU - Ali, Uzma

N1 - This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

PY - 2022/11

Y1 - 2022/11

N2 - Anesthesia drug overdose hazards and lack of gold standards in anesthesia monitoring lead to an urgent need for accurate anesthesia drug detection. To investigate the PPG waveform features affected by anesthesia drugs and develop a machine-learning classifier with high anesthesia drug sensitivity. This study used 64 anesthesia and non-anesthesia patient data (32 cases each), extracted from Queensland and MIMIC-II databases, respectively. The key waveform features (total area, rising time, width 75%, 50%, and 25%) were extracted from 16,310 signal recordings (5-s duration). Discriminant analysis, support vector machine (SVM), and K-nearest neighbor (KNN) were evaluated by splitting the dataset into halve training (11 patients, 8570 segments) and halve testing dataset (11 patients, 7740 segments). Significant differences exist between PPG waveform features of anesthesia and non-anesthesia groups (p 0.05). The KNN classifier achieved 91.7% (AUC = 0.95) anesthesia detection accuracy with the highest sensitivity (0.88) and specificity (0.90) as compared to other classifiers. Kohen's kappa also shows almost perfect agreement (0.79) with the KNN classifier. The KNN classifier trained with significant PPG features has the potential to be used as a reliable, non-invasive, and low-cost method for the detection of anesthesia drugs for depth analysis during surgical operations and postoperative monitoring. [Abstract copyright: © 2022. The Author(s).]

AB - Anesthesia drug overdose hazards and lack of gold standards in anesthesia monitoring lead to an urgent need for accurate anesthesia drug detection. To investigate the PPG waveform features affected by anesthesia drugs and develop a machine-learning classifier with high anesthesia drug sensitivity. This study used 64 anesthesia and non-anesthesia patient data (32 cases each), extracted from Queensland and MIMIC-II databases, respectively. The key waveform features (total area, rising time, width 75%, 50%, and 25%) were extracted from 16,310 signal recordings (5-s duration). Discriminant analysis, support vector machine (SVM), and K-nearest neighbor (KNN) were evaluated by splitting the dataset into halve training (11 patients, 8570 segments) and halve testing dataset (11 patients, 7740 segments). Significant differences exist between PPG waveform features of anesthesia and non-anesthesia groups (p 0.05). The KNN classifier achieved 91.7% (AUC = 0.95) anesthesia detection accuracy with the highest sensitivity (0.88) and specificity (0.90) as compared to other classifiers. Kohen's kappa also shows almost perfect agreement (0.79) with the KNN classifier. The KNN classifier trained with significant PPG features has the potential to be used as a reliable, non-invasive, and low-cost method for the detection of anesthesia drugs for depth analysis during surgical operations and postoperative monitoring. [Abstract copyright: © 2022. The Author(s).]

KW - Queensland database

KW - MIMIC II database

KW - Anesthesia depth

KW - K-nearest neighbor

KW - Photoplethysmography

UR - http://www.scopus.com/inward/record.url?scp=85137419698&partnerID=8YFLogxK

U2 - 10.1007/s11517-022-02658-1

DO - 10.1007/s11517-022-02658-1

M3 - Article

C2 - 36063352

SN - 0140-0118

VL - 60

SP - 3057

EP - 3068

JO - Medical & biological engineering & computing

JF - Medical & biological engineering & computing

IS - 11

ER -

Photoplethysmography temporal marker-based machine learning classifier for anesthesia drug detection

Abstract

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this