Data Driven Investigation of Bispectral Index Algorithm

Hyung Chul Lee, Ho Geol Ryu, Yoonsang Park, Soo Bin Yoon, Seong Mi Yang, Hye Won Oh, Chul Woo Jung

Research output: Contribution to journalArticle

Abstract

Bispectral index (BIS), a useful marker of anaesthetic depth, is calculated by a statistical multivariate model using nonlinear functions of electroencephalography-based subparameters. However, only a portion of the proprietary algorithm has been identified. We investigated the BIS algorithm using clinical big data and machine learning techniques. Retrospective data from 5,427 patients who underwent BIS monitoring during general anaesthesia were used, of which 80% and 20% were used as training datasets and test datasets, respectively. A histogram of data points was plotted to define five BIS ranges representing the depth of anaesthesia. Decision tree analysis was performed to determine the electroencephalography subparameters and their thresholds for classifying five BIS ranges. Random sample consensus regression analyses were performed using the subparameters to derive multiple linear regression models of BIS calculation in five BIS ranges. The performance of the decision tree and regression models was externally validated with positive predictive value and median absolute error, respectively. A four-level depth decision tree was built with four subparameters such as burst suppression ratio, power of electromyogram, 95% spectral edge frequency, and relative beta ratio. Positive predictive values were 100%, 80%, 80%, 85% and 89% in the order of increasing BIS in the five BIS ranges. The average of median absolute errors of regression models was 4.1 as BIS value. A data driven BIS calculation algorithm using multiple electroencephalography subparameters with different weights depending on BIS ranges has been proposed. The results may help the anaesthesiologists interpret the erroneous BIS values observed during clinical practice.

Original languageEnglish
Article number13769
JournalScientific Reports
Volume9
Issue number1
DOIs
StatePublished - 1 Dec 2019

Fingerprint

Decision Trees
Electroencephalography
Linear Models
Decision Support Techniques
Electromyography
Statistical Models
General Anesthesia
Anesthetics
Anesthesia
Regression Analysis
Weights and Measures
Datasets

Cite this

@article{dcd71e792a3c47069984b7c937dfb4b2,
title = "Data Driven Investigation of Bispectral Index Algorithm",
abstract = "Bispectral index (BIS), a useful marker of anaesthetic depth, is calculated by a statistical multivariate model using nonlinear functions of electroencephalography-based subparameters. However, only a portion of the proprietary algorithm has been identified. We investigated the BIS algorithm using clinical big data and machine learning techniques. Retrospective data from 5,427 patients who underwent BIS monitoring during general anaesthesia were used, of which 80{\%} and 20{\%} were used as training datasets and test datasets, respectively. A histogram of data points was plotted to define five BIS ranges representing the depth of anaesthesia. Decision tree analysis was performed to determine the electroencephalography subparameters and their thresholds for classifying five BIS ranges. Random sample consensus regression analyses were performed using the subparameters to derive multiple linear regression models of BIS calculation in five BIS ranges. The performance of the decision tree and regression models was externally validated with positive predictive value and median absolute error, respectively. A four-level depth decision tree was built with four subparameters such as burst suppression ratio, power of electromyogram, 95{\%} spectral edge frequency, and relative beta ratio. Positive predictive values were 100{\%}, 80{\%}, 80{\%}, 85{\%} and 89{\%} in the order of increasing BIS in the five BIS ranges. The average of median absolute errors of regression models was 4.1 as BIS value. A data driven BIS calculation algorithm using multiple electroencephalography subparameters with different weights depending on BIS ranges has been proposed. The results may help the anaesthesiologists interpret the erroneous BIS values observed during clinical practice.",
author = "Lee, {Hyung Chul} and Ryu, {Ho Geol} and Yoonsang Park and Yoon, {Soo Bin} and Yang, {Seong Mi} and Oh, {Hye Won} and Jung, {Chul Woo}",
year = "2019",
month = "12",
day = "1",
doi = "10.1038/s41598-019-50391-x",
language = "English",
volume = "9",
journal = "Scientific reports",
issn = "2045-2322",
publisher = "Nature Publishing Group",
number = "1",

}

Data Driven Investigation of Bispectral Index Algorithm. / Lee, Hyung Chul; Ryu, Ho Geol; Park, Yoonsang; Yoon, Soo Bin; Yang, Seong Mi; Oh, Hye Won; Jung, Chul Woo.

In: Scientific Reports, Vol. 9, No. 1, 13769, 01.12.2019.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Data Driven Investigation of Bispectral Index Algorithm

AU - Lee, Hyung Chul

AU - Ryu, Ho Geol

AU - Park, Yoonsang

AU - Yoon, Soo Bin

AU - Yang, Seong Mi

AU - Oh, Hye Won

AU - Jung, Chul Woo

PY - 2019/12/1

Y1 - 2019/12/1

N2 - Bispectral index (BIS), a useful marker of anaesthetic depth, is calculated by a statistical multivariate model using nonlinear functions of electroencephalography-based subparameters. However, only a portion of the proprietary algorithm has been identified. We investigated the BIS algorithm using clinical big data and machine learning techniques. Retrospective data from 5,427 patients who underwent BIS monitoring during general anaesthesia were used, of which 80% and 20% were used as training datasets and test datasets, respectively. A histogram of data points was plotted to define five BIS ranges representing the depth of anaesthesia. Decision tree analysis was performed to determine the electroencephalography subparameters and their thresholds for classifying five BIS ranges. Random sample consensus regression analyses were performed using the subparameters to derive multiple linear regression models of BIS calculation in five BIS ranges. The performance of the decision tree and regression models was externally validated with positive predictive value and median absolute error, respectively. A four-level depth decision tree was built with four subparameters such as burst suppression ratio, power of electromyogram, 95% spectral edge frequency, and relative beta ratio. Positive predictive values were 100%, 80%, 80%, 85% and 89% in the order of increasing BIS in the five BIS ranges. The average of median absolute errors of regression models was 4.1 as BIS value. A data driven BIS calculation algorithm using multiple electroencephalography subparameters with different weights depending on BIS ranges has been proposed. The results may help the anaesthesiologists interpret the erroneous BIS values observed during clinical practice.

AB - Bispectral index (BIS), a useful marker of anaesthetic depth, is calculated by a statistical multivariate model using nonlinear functions of electroencephalography-based subparameters. However, only a portion of the proprietary algorithm has been identified. We investigated the BIS algorithm using clinical big data and machine learning techniques. Retrospective data from 5,427 patients who underwent BIS monitoring during general anaesthesia were used, of which 80% and 20% were used as training datasets and test datasets, respectively. A histogram of data points was plotted to define five BIS ranges representing the depth of anaesthesia. Decision tree analysis was performed to determine the electroencephalography subparameters and their thresholds for classifying five BIS ranges. Random sample consensus regression analyses were performed using the subparameters to derive multiple linear regression models of BIS calculation in five BIS ranges. The performance of the decision tree and regression models was externally validated with positive predictive value and median absolute error, respectively. A four-level depth decision tree was built with four subparameters such as burst suppression ratio, power of electromyogram, 95% spectral edge frequency, and relative beta ratio. Positive predictive values were 100%, 80%, 80%, 85% and 89% in the order of increasing BIS in the five BIS ranges. The average of median absolute errors of regression models was 4.1 as BIS value. A data driven BIS calculation algorithm using multiple electroencephalography subparameters with different weights depending on BIS ranges has been proposed. The results may help the anaesthesiologists interpret the erroneous BIS values observed during clinical practice.

UR - http://www.scopus.com/inward/record.url?scp=85072595789&partnerID=8YFLogxK

U2 - 10.1038/s41598-019-50391-x

DO - 10.1038/s41598-019-50391-x

M3 - Article

C2 - 31551487

AN - SCOPUS:85072595789

VL - 9

JO - Scientific reports

JF - Scientific reports

SN - 2045-2322

IS - 1

M1 - 13769

ER -