AI-Enabled ECG Combined with Dry Electrode Sensors for Population-Based Screening of Atrial Fibrillation

Alan Kennedy; Dewar D. Finlay; Raymond Bond; Daniel Guldenring; James McLaughlin; Chris Crockford

doi:10.22489/CinC.2022.312

AI-Enabled ECG Combined with Dry Electrode Sensors for Population-Based Screening of Atrial Fibrillation

Alan Kennedy, Dewar D. Finlay, Raymond Bond, Daniel Guldenring, James McLaughlin, Chris Crockford

Faculty Of Computing, Eng. & Built Env.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

This study assessed the performance of a deep neural network (PulseAI, Belfast, United Kingdom) used in conjunction with a dry-electrode ECG sensor device (RhythmPad, D&FT, United Kingdom) to detect AF automatically. Simultaneous pairs of 12-lead ECGs and single-lead dry-electrode ECGs were collected from 622 patients. The 12-lead ECGs were manually overread and used as reference diagnoses. Twenty-two patients were confirmed with AF and had an interpretable 12-lead and single-lead dry-electrode ECG recording. The deep neural network analysed the dry-electrode ECGs, and performance was compared to the 12-lead interpretation. Overall, the deep neural network algorithm yielded a sensitivity of 96% (95% CI, 87%-100%), specificity of 99% (95% CI, 98%-100%) and positive predictive value of 81% (95% CI, 66%-96%) for detection of AF episodes. When coupled with dry-electrode ECG sensors, the PulseAI neural network allows for large-scale and low-cost screening for AF. Widespread implementation of this technology may allow for earlier detection, treatment, and management of patients with AF.

Original language	English
Title of host publication	2022 Computing in Cardiology, CinC 2022
Publisher	IEEE Computer Society
Pages	1-4
Number of pages	4
Volume	49
ISBN (Electronic)	9798350300970
DOIs	https://doi.org/10.22489/CinC.2022.312
Publication status	Published online - 3 Apr 2023
Event	2022 Computing in Cardiology, CinC 2022 - Tampere, Finland Duration: 4 Sept 2022 → 7 Sept 2022

Publication series

Name	Computing in Cardiology
Volume	2022-September
ISSN (Print)	2325-8861
ISSN (Electronic)	2325-887X

Conference

Conference	2022 Computing in Cardiology, CinC 2022
Country/Territory	Finland
City	Tampere
Period	4/09/22 → 7/09/22

Bibliographical note

Funding Information:
Funding to present this research was provided by the European Union's INTERREG VA Programme, managed by the Special EU Programmes Body (SEUPB), which is part of the Eastern Corridor Medical Engineering centre (ECME).

Publisher Copyright:
© 2022 Creative Commons.

Keywords

Deep learning
Performance evaluation
Electrodes
Sensitivity
Neural networks
Europe
Electrocardiography

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.22489/CinC.2022.312

Cite this

Kennedy, A., Finlay, D. D., Bond, R., Guldenring, D., McLaughlin, J., & Crockford, C. (2023). AI-Enabled ECG Combined with Dry Electrode Sensors for Population-Based Screening of Atrial Fibrillation. In 2022 Computing in Cardiology, CinC 2022 (Vol. 49, pp. 1-4). (Computing in Cardiology; Vol. 2022-September). IEEE Computer Society. Advance online publication. https://doi.org/10.22489/CinC.2022.312

@inproceedings{a90273f2b1ca4e54a10c401352159e89,

title = "AI-Enabled ECG Combined with Dry Electrode Sensors for Population-Based Screening of Atrial Fibrillation",

abstract = "This study assessed the performance of a deep neural network (PulseAI, Belfast, United Kingdom) used in conjunction with a dry-electrode ECG sensor device (RhythmPad, D&FT, United Kingdom) to detect AF automatically. Simultaneous pairs of 12-lead ECGs and single-lead dry-electrode ECGs were collected from 622 patients. The 12-lead ECGs were manually overread and used as reference diagnoses. Twenty-two patients were confirmed with AF and had an interpretable 12-lead and single-lead dry-electrode ECG recording. The deep neural network analysed the dry-electrode ECGs, and performance was compared to the 12-lead interpretation. Overall, the deep neural network algorithm yielded a sensitivity of 96% (95% CI, 87%-100%), specificity of 99% (95% CI, 98%-100%) and positive predictive value of 81% (95% CI, 66%-96%) for detection of AF episodes. When coupled with dry-electrode ECG sensors, the PulseAI neural network allows for large-scale and low-cost screening for AF. Widespread implementation of this technology may allow for earlier detection, treatment, and management of patients with AF.",

keywords = "Deep learning, Performance evaluation, Electrodes, Sensitivity, Neural networks, Europe, Electrocardiography",

author = "Alan Kennedy and Finlay, {Dewar D.} and Raymond Bond and Daniel Guldenring and James McLaughlin and Chris Crockford",

note = "Funding Information: Funding to present this research was provided by the European Union's INTERREG VA Programme, managed by the Special EU Programmes Body (SEUPB), which is part of the Eastern Corridor Medical Engineering centre (ECME). Publisher Copyright: {\textcopyright} 2022 Creative Commons.; 2022 Computing in Cardiology, CinC 2022 ; Conference date: 04-09-2022 Through 07-09-2022",

year = "2023",

month = apr,

day = "3",

doi = "10.22489/CinC.2022.312",

language = "English",

volume = "49",

series = "Computing in Cardiology",

publisher = "IEEE Computer Society",

pages = "1--4",

booktitle = "2022 Computing in Cardiology, CinC 2022",

address = "United States",

}

Kennedy, A, Finlay, DD , Bond, R, Guldenring, D, McLaughlin, J & Crockford, C 2023, AI-Enabled ECG Combined with Dry Electrode Sensors for Population-Based Screening of Atrial Fibrillation. in 2022 Computing in Cardiology, CinC 2022. vol. 49, Computing in Cardiology, vol. 2022-September, IEEE Computer Society, pp. 1-4, 2022 Computing in Cardiology, CinC 2022, Tampere, Finland, 4/09/22. https://doi.org/10.22489/CinC.2022.312

AI-Enabled ECG Combined with Dry Electrode Sensors for Population-Based Screening of Atrial Fibrillation. / Kennedy, Alan; Finlay, Dewar D.; Bond, Raymond et al.
2022 Computing in Cardiology, CinC 2022. Vol. 49 IEEE Computer Society, 2023. p. 1-4 (Computing in Cardiology; Vol. 2022-September).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - AI-Enabled ECG Combined with Dry Electrode Sensors for Population-Based Screening of Atrial Fibrillation

AU - Kennedy, Alan

AU - Finlay, Dewar D.

AU - Bond, Raymond

AU - Guldenring, Daniel

AU - McLaughlin, James

AU - Crockford, Chris

N1 - Funding Information: Funding to present this research was provided by the European Union's INTERREG VA Programme, managed by the Special EU Programmes Body (SEUPB), which is part of the Eastern Corridor Medical Engineering centre (ECME). Publisher Copyright: © 2022 Creative Commons.

PY - 2023/4/3

Y1 - 2023/4/3

N2 - This study assessed the performance of a deep neural network (PulseAI, Belfast, United Kingdom) used in conjunction with a dry-electrode ECG sensor device (RhythmPad, D&FT, United Kingdom) to detect AF automatically. Simultaneous pairs of 12-lead ECGs and single-lead dry-electrode ECGs were collected from 622 patients. The 12-lead ECGs were manually overread and used as reference diagnoses. Twenty-two patients were confirmed with AF and had an interpretable 12-lead and single-lead dry-electrode ECG recording. The deep neural network analysed the dry-electrode ECGs, and performance was compared to the 12-lead interpretation. Overall, the deep neural network algorithm yielded a sensitivity of 96% (95% CI, 87%-100%), specificity of 99% (95% CI, 98%-100%) and positive predictive value of 81% (95% CI, 66%-96%) for detection of AF episodes. When coupled with dry-electrode ECG sensors, the PulseAI neural network allows for large-scale and low-cost screening for AF. Widespread implementation of this technology may allow for earlier detection, treatment, and management of patients with AF.

AB - This study assessed the performance of a deep neural network (PulseAI, Belfast, United Kingdom) used in conjunction with a dry-electrode ECG sensor device (RhythmPad, D&FT, United Kingdom) to detect AF automatically. Simultaneous pairs of 12-lead ECGs and single-lead dry-electrode ECGs were collected from 622 patients. The 12-lead ECGs were manually overread and used as reference diagnoses. Twenty-two patients were confirmed with AF and had an interpretable 12-lead and single-lead dry-electrode ECG recording. The deep neural network analysed the dry-electrode ECGs, and performance was compared to the 12-lead interpretation. Overall, the deep neural network algorithm yielded a sensitivity of 96% (95% CI, 87%-100%), specificity of 99% (95% CI, 98%-100%) and positive predictive value of 81% (95% CI, 66%-96%) for detection of AF episodes. When coupled with dry-electrode ECG sensors, the PulseAI neural network allows for large-scale and low-cost screening for AF. Widespread implementation of this technology may allow for earlier detection, treatment, and management of patients with AF.

KW - Deep learning

KW - Performance evaluation

KW - Electrodes

KW - Sensitivity

KW - Neural networks

KW - Europe

KW - Electrocardiography

UR - http://www.scopus.com/inward/record.url?scp=85152887352&partnerID=8YFLogxK

UR - https://ieeexplore.ieee.org/document/10081844

U2 - 10.22489/CinC.2022.312

DO - 10.22489/CinC.2022.312

M3 - Conference contribution

AN - SCOPUS:85152887352

VL - 49

T3 - Computing in Cardiology

SP - 1

EP - 4

BT - 2022 Computing in Cardiology, CinC 2022

PB - IEEE Computer Society

T2 - 2022 Computing in Cardiology, CinC 2022

Y2 - 4 September 2022 through 7 September 2022

ER -

AI-Enabled ECG Combined with Dry Electrode Sensors for Population-Based Screening of Atrial Fibrillation

Abstract

Publication series

Conference

Bibliographical note

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this