Explaining deep neural networks for knowledge discovery in electrocardiogram analysis

Research output: Contribution to journal › Journal article › Research › peer-review

Standard

Explaining deep neural networks for knowledge discovery in electrocardiogram analysis. / Hicks, Steven A; Isaksen, Jonas L; Thambawita, Vajira; Ghouse, Jonas; Ahlberg, Gustav; Linneberg, Allan; Grarup, Niels; Strümke, Inga; Ellervik, Christina; Olesen, Morten Salling; Hansen, Torben; Graff, Claus; Holstein-Rathlou, Niels-Henrik; Halvorsen, Pål; Maleckar, Mary M; Riegler, Michael A; Kanters, Jørgen K.

In: Scientific Reports, Vol. 11, No. 1, 10949, 2021.

Research output: Contribution to journal › Journal article › Research › peer-review

Harvard

Hicks, SA, Isaksen, JL, Thambawita, V, Ghouse, J, Ahlberg, G, Linneberg, A, Grarup, N, Strümke, I, Ellervik, C, Olesen, MS, Hansen, T, Graff, C, Holstein-Rathlou, N-H, Halvorsen, P, Maleckar, MM, Riegler, MA & Kanters, JK 2021, 'Explaining deep neural networks for knowledge discovery in electrocardiogram analysis', Scientific Reports, vol. 11, no. 1, 10949. https://doi.org/10.1038/s41598-021-90285-5

APA

Hicks, S. A., Isaksen, J. L., Thambawita, V., Ghouse, J., Ahlberg, G., Linneberg, A., Grarup, N., Strümke, I., Ellervik, C., Olesen, M. S., Hansen, T., Graff, C., Holstein-Rathlou, N-H., Halvorsen, P., Maleckar, M. M., Riegler, M. A., & Kanters, J. K. (2021). Explaining deep neural networks for knowledge discovery in electrocardiogram analysis. Scientific Reports, 11(1), [10949]. https://doi.org/10.1038/s41598-021-90285-5

Vancouver

Hicks SA, Isaksen JL, Thambawita V, Ghouse J, Ahlberg G, Linneberg A et al. Explaining deep neural networks for knowledge discovery in electrocardiogram analysis. Scientific Reports. 2021;11(1). 10949. https://doi.org/10.1038/s41598-021-90285-5

Author

Hicks, Steven A ; Isaksen, Jonas L ; Thambawita, Vajira ; Ghouse, Jonas ; Ahlberg, Gustav ; Linneberg, Allan ; Grarup, Niels ; Strümke, Inga ; Ellervik, Christina ; Olesen, Morten Salling ; Hansen, Torben ; Graff, Claus ; Holstein-Rathlou, Niels-Henrik ; Halvorsen, Pål ; Maleckar, Mary M ; Riegler, Michael A ; Kanters, Jørgen K. / Explaining deep neural networks for knowledge discovery in electrocardiogram analysis. In: Scientific Reports. 2021 ; Vol. 11, No. 1.

Bibtex

@article{cfb6e91e081d465fadedca4658b3ff07,
title = "Explaining deep neural networks for knowledge discovery in electrocardiogram analysis",
abstract = "Deep learning-based tools may annotate and interpret medical data more quickly, consistently, and accurately than medical doctors. However, as medical doctors are ultimately responsible for clinical decision-making, any deep learning-based prediction should be accompanied by an explanation that a human can understand. We present an approach called electrocardiogram gradient class activation map (ECGradCAM), which is used to generate attention maps and explain the reasoning behind deep learning-based decision-making in ECG analysis. Attention maps may be used in the clinic to aid diagnosis, discover new medical knowledge, and identify novel features and characteristics of medical tests. In this paper, we showcase how ECGradCAM attention maps can unmask how a novel deep learning model measures both amplitudes and intervals in 12-lead electrocardiograms, and we show an example of how attention maps may be used to develop novel ECG features.",
author = "Hicks, {Steven A} and Isaksen, {Jonas L} and Vajira Thambawita and Jonas Ghouse and Gustav Ahlberg and Allan Linneberg and Niels Grarup and Inga Str{\"u}mke and Christina Ellervik and Olesen, {Morten Salling} and Torben Hansen and Claus Graff and Niels-Henrik Holstein-Rathlou and P{\aa}l Halvorsen and Maleckar, {Mary M} and Riegler, {Michael A} and Kanters, {J{\o}rgen K}",
year = "2021",
doi = "10.1038/s41598-021-90285-5",
language = "English",
volume = "11",
journal = "Scientific Reports",
issn = "2045-2322",
publisher = "nature publishing group",
number = "1",

}

RIS

TY - JOUR

T1 - Explaining deep neural networks for knowledge discovery in electrocardiogram analysis

AU - Hicks, Steven A

AU - Isaksen, Jonas L

AU - Thambawita, Vajira

AU - Ghouse, Jonas

AU - Ahlberg, Gustav

AU - Linneberg, Allan

AU - Grarup, Niels

AU - Strümke, Inga

AU - Ellervik, Christina

AU - Olesen, Morten Salling

AU - Hansen, Torben

AU - Graff, Claus

AU - Holstein-Rathlou, Niels-Henrik

AU - Halvorsen, Pål

AU - Maleckar, Mary M

AU - Riegler, Michael A

AU - Kanters, Jørgen K

PY - 2021

Y1 - 2021

N2 - Deep learning-based tools may annotate and interpret medical data more quickly, consistently, and accurately than medical doctors. However, as medical doctors are ultimately responsible for clinical decision-making, any deep learning-based prediction should be accompanied by an explanation that a human can understand. We present an approach called electrocardiogram gradient class activation map (ECGradCAM), which is used to generate attention maps and explain the reasoning behind deep learning-based decision-making in ECG analysis. Attention maps may be used in the clinic to aid diagnosis, discover new medical knowledge, and identify novel features and characteristics of medical tests. In this paper, we showcase how ECGradCAM attention maps can unmask how a novel deep learning model measures both amplitudes and intervals in 12-lead electrocardiograms, and we show an example of how attention maps may be used to develop novel ECG features.

AB - Deep learning-based tools may annotate and interpret medical data more quickly, consistently, and accurately than medical doctors. However, as medical doctors are ultimately responsible for clinical decision-making, any deep learning-based prediction should be accompanied by an explanation that a human can understand. We present an approach called electrocardiogram gradient class activation map (ECGradCAM), which is used to generate attention maps and explain the reasoning behind deep learning-based decision-making in ECG analysis. Attention maps may be used in the clinic to aid diagnosis, discover new medical knowledge, and identify novel features and characteristics of medical tests. In this paper, we showcase how ECGradCAM attention maps can unmask how a novel deep learning model measures both amplitudes and intervals in 12-lead electrocardiograms, and we show an example of how attention maps may be used to develop novel ECG features.

U2 - 10.1038/s41598-021-90285-5

DO - 10.1038/s41598-021-90285-5

M3 - Journal article

C2 - 34040033

VL - 11

JO - Scientific Reports

JF - Scientific Reports

SN - 2045-2322

IS - 1

M1 - 10949

ER -

ID: 269904130