Symbolic dynamics to enhance diagnostic ability of portable oximetry from the Phone Oximeter in the detection of paediatric sleep apnoea

Daniel Alvarez (Corresponding Author), Andrea Crespo, Fernando Vaquerizo-Villar, Gonzalo C. Gutierrez-Tobal, Ana Cerezo-Hernandez, Veronica Barroso-Garcia, J. Mark Ansermino, Guy A. Dumont, Roberto Hornero, Felix del Campo, Ainara Garde

Research output: Contribution to journalArticleAcademicpeer-review

Abstract

Objective: This study is aimed at assessing symbolic dynamics as a reliable technique to characterize complex fluctuations of portable oximetry in the context of automated detection of childhood obstructive sleep apnoea-hypopnoea syndrome (OSAHS). Approach: Nocturnal oximetry signals from 142 children with suspected OSAHS were acquired using the Phone Oximeter: a portable device that integrates a pulse oximeter with a smartphone. An apnoea-hypopnoea index (AHI) ⩾ 5 events h−1 from simultaneous in-lab polysomnography was used to confirm moderate-to-severe childhood OSAHS. Symbolic dynamics was used to parameterise non-linear changes in the overnight oximetry profile. Conventional indices, anthropometric measures, and time-domain linear statistics were also considered. Forward stepwise logistic regression was used to obtain an optimum feature subset. Logistic regression (LR) was used to identify children with moderate-to-severe OSAHS. Main results: The histogram of 3-symbol words from symbolic dynamics showed significant differences (p < 0.01) between children with AHI < 5 events h−1 and moderate-to-severe patients (AHI ⩾ 5 events h−1). Words representing increasing oximetry values after apnoeic events (re-saturations) showed relevant diagnostic information. Regarding the performance of individual characterization approaches, the LR model composed of features from symbolic dynamics alone reached a maximum performance of 78.4% accuracy (65.2% sensitivity; 86.8% specificity) and 0.83 area under the ROC curve (AUC). The classification performance improved combining all features. The optimum model from feature selection achieved 83.3% accuracy (73.5% sensitivity; 89.5% specificity) and 0.89 AUC, significantly (p <0.01) outperforming the other models. Significance: Symbolic dynamics provides complementary information to conventional oximetry analysis enabling reliable detection of moderate-to-severe paediatric OSAHS from portable oximetry.
Original languageEnglish
Article number104002
Number of pages16
JournalPhysiological measurement
Volume39
Issue number10
Early online date19 Sep 2018
DOIs
Publication statusPublished - 11 Oct 2018

Keywords

  • paediatric obstructive sleep apnoea-hypopnoea syndrome
  • nocturnal oximetry
  • portable
  • signal processing
  • symbolic dynamics
  • pattern recognition

Cite this

Alvarez, Daniel ; Crespo, Andrea ; Vaquerizo-Villar, Fernando ; Gutierrez-Tobal, Gonzalo C. ; Cerezo-Hernandez, Ana ; Barroso-Garcia, Veronica ; Mark Ansermino, J. ; Dumont, Guy A. ; Hornero, Roberto ; del Campo, Felix ; Garde, Ainara. / Symbolic dynamics to enhance diagnostic ability of portable oximetry from the Phone Oximeter in the detection of paediatric sleep apnoea. In: Physiological measurement. 2018 ; Vol. 39, No. 10.
@article{cc4f51789ff04b75be15f203bc4071a1,
title = "Symbolic dynamics to enhance diagnostic ability of portable oximetry from the Phone Oximeter in the detection of paediatric sleep apnoea",
abstract = "Objective: This study is aimed at assessing symbolic dynamics as a reliable technique to characterize complex fluctuations of portable oximetry in the context of automated detection of childhood obstructive sleep apnoea-hypopnoea syndrome (OSAHS). Approach: Nocturnal oximetry signals from 142 children with suspected OSAHS were acquired using the Phone Oximeter: a portable device that integrates a pulse oximeter with a smartphone. An apnoea-hypopnoea index (AHI) ⩾ 5 events h−1 from simultaneous in-lab polysomnography was used to confirm moderate-to-severe childhood OSAHS. Symbolic dynamics was used to parameterise non-linear changes in the overnight oximetry profile. Conventional indices, anthropometric measures, and time-domain linear statistics were also considered. Forward stepwise logistic regression was used to obtain an optimum feature subset. Logistic regression (LR) was used to identify children with moderate-to-severe OSAHS. Main results: The histogram of 3-symbol words from symbolic dynamics showed significant differences (p < 0.01) between children with AHI < 5 events h−1 and moderate-to-severe patients (AHI ⩾ 5 events h−1). Words representing increasing oximetry values after apnoeic events (re-saturations) showed relevant diagnostic information. Regarding the performance of individual characterization approaches, the LR model composed of features from symbolic dynamics alone reached a maximum performance of 78.4{\%} accuracy (65.2{\%} sensitivity; 86.8{\%} specificity) and 0.83 area under the ROC curve (AUC). The classification performance improved combining all features. The optimum model from feature selection achieved 83.3{\%} accuracy (73.5{\%} sensitivity; 89.5{\%} specificity) and 0.89 AUC, significantly (p <0.01) outperforming the other models. Significance: Symbolic dynamics provides complementary information to conventional oximetry analysis enabling reliable detection of moderate-to-severe paediatric OSAHS from portable oximetry.",
keywords = "paediatric obstructive sleep apnoea-hypopnoea syndrome, nocturnal oximetry, portable, signal processing, symbolic dynamics, pattern recognition",
author = "Daniel Alvarez and Andrea Crespo and Fernando Vaquerizo-Villar and Gutierrez-Tobal, {Gonzalo C.} and Ana Cerezo-Hernandez and Veronica Barroso-Garcia and {Mark Ansermino}, J. and Dumont, {Guy A.} and Roberto Hornero and {del Campo}, Felix and Ainara Garde",
year = "2018",
month = "10",
day = "11",
doi = "10.1088/1361-6579/aae2a8",
language = "English",
volume = "39",
journal = "Physiological measurement",
issn = "0967-3334",
publisher = "IOP Publishing Ltd.",
number = "10",

}

Alvarez, D, Crespo, A, Vaquerizo-Villar, F, Gutierrez-Tobal, GC, Cerezo-Hernandez, A, Barroso-Garcia, V, Mark Ansermino, J, Dumont, GA, Hornero, R, del Campo, F & Garde, A 2018, 'Symbolic dynamics to enhance diagnostic ability of portable oximetry from the Phone Oximeter in the detection of paediatric sleep apnoea' Physiological measurement, vol. 39, no. 10, 104002. https://doi.org/10.1088/1361-6579/aae2a8

Symbolic dynamics to enhance diagnostic ability of portable oximetry from the Phone Oximeter in the detection of paediatric sleep apnoea. / Alvarez, Daniel (Corresponding Author); Crespo, Andrea; Vaquerizo-Villar, Fernando; Gutierrez-Tobal, Gonzalo C.; Cerezo-Hernandez, Ana; Barroso-Garcia, Veronica; Mark Ansermino, J.; Dumont, Guy A.; Hornero, Roberto; del Campo, Felix; Garde, Ainara.

In: Physiological measurement, Vol. 39, No. 10, 104002, 11.10.2018.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - Symbolic dynamics to enhance diagnostic ability of portable oximetry from the Phone Oximeter in the detection of paediatric sleep apnoea

AU - Alvarez, Daniel

AU - Crespo, Andrea

AU - Vaquerizo-Villar, Fernando

AU - Gutierrez-Tobal, Gonzalo C.

AU - Cerezo-Hernandez, Ana

AU - Barroso-Garcia, Veronica

AU - Mark Ansermino, J.

AU - Dumont, Guy A.

AU - Hornero, Roberto

AU - del Campo, Felix

AU - Garde, Ainara

PY - 2018/10/11

Y1 - 2018/10/11

N2 - Objective: This study is aimed at assessing symbolic dynamics as a reliable technique to characterize complex fluctuations of portable oximetry in the context of automated detection of childhood obstructive sleep apnoea-hypopnoea syndrome (OSAHS). Approach: Nocturnal oximetry signals from 142 children with suspected OSAHS were acquired using the Phone Oximeter: a portable device that integrates a pulse oximeter with a smartphone. An apnoea-hypopnoea index (AHI) ⩾ 5 events h−1 from simultaneous in-lab polysomnography was used to confirm moderate-to-severe childhood OSAHS. Symbolic dynamics was used to parameterise non-linear changes in the overnight oximetry profile. Conventional indices, anthropometric measures, and time-domain linear statistics were also considered. Forward stepwise logistic regression was used to obtain an optimum feature subset. Logistic regression (LR) was used to identify children with moderate-to-severe OSAHS. Main results: The histogram of 3-symbol words from symbolic dynamics showed significant differences (p < 0.01) between children with AHI < 5 events h−1 and moderate-to-severe patients (AHI ⩾ 5 events h−1). Words representing increasing oximetry values after apnoeic events (re-saturations) showed relevant diagnostic information. Regarding the performance of individual characterization approaches, the LR model composed of features from symbolic dynamics alone reached a maximum performance of 78.4% accuracy (65.2% sensitivity; 86.8% specificity) and 0.83 area under the ROC curve (AUC). The classification performance improved combining all features. The optimum model from feature selection achieved 83.3% accuracy (73.5% sensitivity; 89.5% specificity) and 0.89 AUC, significantly (p <0.01) outperforming the other models. Significance: Symbolic dynamics provides complementary information to conventional oximetry analysis enabling reliable detection of moderate-to-severe paediatric OSAHS from portable oximetry.

AB - Objective: This study is aimed at assessing symbolic dynamics as a reliable technique to characterize complex fluctuations of portable oximetry in the context of automated detection of childhood obstructive sleep apnoea-hypopnoea syndrome (OSAHS). Approach: Nocturnal oximetry signals from 142 children with suspected OSAHS were acquired using the Phone Oximeter: a portable device that integrates a pulse oximeter with a smartphone. An apnoea-hypopnoea index (AHI) ⩾ 5 events h−1 from simultaneous in-lab polysomnography was used to confirm moderate-to-severe childhood OSAHS. Symbolic dynamics was used to parameterise non-linear changes in the overnight oximetry profile. Conventional indices, anthropometric measures, and time-domain linear statistics were also considered. Forward stepwise logistic regression was used to obtain an optimum feature subset. Logistic regression (LR) was used to identify children with moderate-to-severe OSAHS. Main results: The histogram of 3-symbol words from symbolic dynamics showed significant differences (p < 0.01) between children with AHI < 5 events h−1 and moderate-to-severe patients (AHI ⩾ 5 events h−1). Words representing increasing oximetry values after apnoeic events (re-saturations) showed relevant diagnostic information. Regarding the performance of individual characterization approaches, the LR model composed of features from symbolic dynamics alone reached a maximum performance of 78.4% accuracy (65.2% sensitivity; 86.8% specificity) and 0.83 area under the ROC curve (AUC). The classification performance improved combining all features. The optimum model from feature selection achieved 83.3% accuracy (73.5% sensitivity; 89.5% specificity) and 0.89 AUC, significantly (p <0.01) outperforming the other models. Significance: Symbolic dynamics provides complementary information to conventional oximetry analysis enabling reliable detection of moderate-to-severe paediatric OSAHS from portable oximetry.

KW - paediatric obstructive sleep apnoea-hypopnoea syndrome

KW - nocturnal oximetry

KW - portable

KW - signal processing

KW - symbolic dynamics

KW - pattern recognition

U2 - 10.1088/1361-6579/aae2a8

DO - 10.1088/1361-6579/aae2a8

M3 - Article

VL - 39

JO - Physiological measurement

JF - Physiological measurement

SN - 0967-3334

IS - 10

M1 - 104002

ER -

Alvarez D, Crespo A, Vaquerizo-Villar F, Gutierrez-Tobal GC, Cerezo-Hernandez A, Barroso-Garcia V et al. Symbolic dynamics to enhance diagnostic ability of portable oximetry from the Phone Oximeter in the detection of paediatric sleep apnoea. Physiological measurement. 2018 Oct 11;39(10). 104002. https://doi.org/10.1088/1361-6579/aae2a8