Uncertainty quantification in DenseNet model using myocardial infarction ECG signals

V Jahmunah; E Y K Ng; Ru-San Tan; Shu Lih Oh; U Rajendra Acharya

doi:10.1016/j.cmpb.2022.107308

Uncertainty quantification in DenseNet model using myocardial infarction ECG signals

Comput Methods Programs Biomed. 2023 Feb:229:107308. doi: 10.1016/j.cmpb.2022.107308. Epub 2022 Dec 12.

Authors

V Jahmunah¹, E Y K Ng¹, Ru-San Tan², Shu Lih Oh³, U Rajendra Acharya⁴

Affiliations

¹ School of Mechanical and Aerospace Engineering, Nanyang Technological University.
² National Heart Centre, Singapore.
³ School of Engineering, Ngee Ann Polytechnic, Singapore.
⁴ School of Engineering, Ngee Ann Polytechnic, Singapore; Biomedical Engineering, School of Social Science and Technology, Singapore University of Social Sciences, Singapore; International Research Organization for Advanced Science and Technology (IROAST), Kumamoto University, Kumamoto, Japan; Department Bioinformatics and Medical Engineering, Asia University, Taiwan.; School of Management and Enterprise, University of Southern Queensland, Australia. Electronic address: aru@np.edu.sg.

PMID: 36535127
DOI: 10.1016/j.cmpb.2022.107308

Abstract

Background and objective: Myocardial infarction (MI) is a life-threatening condition diagnosed acutely on the electrocardiogram (ECG). Several errors, such as noise, can impair the prediction of automated ECG diagnosis. Therefore, quantification and communication of model uncertainty are essential for reliable MI diagnosis.

Methods: A Dirichlet DenseNet model that could analyze out-of-distribution data and detect misclassification of MI and normal ECG signals was developed. The DenseNet model was first trained with the pre-processed MI ECG signals (from the best lead V6) acquired from the Physikalisch-Technische Bundesanstalt (PTB) database, using the reverse Kullback-Leibler (KL) divergence loss. The model was then tested with newly synthesized ECG signals with added em and ma noise samples. Predictive entropy was used as an uncertainty measure to determine the misclassification of normal and MI signals. Model performance was evaluated using four uncertainty metrics: uncertainty sensitivity (UNSE), uncertainty specificity (UNSP), uncertainty accuracy (UNAC), and uncertainty precision (UNPR); the classification threshold was set at 0.3.

Results: The UNSE of the DenseNet model was low but increased over the studied decremental noise range (-6 to 24 dB), indicating that the model grew more confident in classifying the signals as they got less noisy. The model became more certain in its predictions from SNR values of 12 dB and 18 dB onwards, yielding UNAC values of 80% and 82.4% for em and ma noise signals, respectively. UNSP and UNPR values were close to 100% for em and ma noise signals, indicating that the model was self-aware of what it knew and didn't.

Conclusion: Through this work, it has been established that the model is reliable as it was able to convey when it was not confident in the diagnostic information it was presenting. Thus, the model is trustworthy and can be used in healthcare applications, such as the emergency diagnosis of MI on ECGs.

Keywords: Deep learning; DenseNet model; Myocardial infarction; Predictive entropy; Reverse KL divergence; Uncertainty quantification.

MeSH terms

Databases, Factual
Electrocardiography*
Entropy
Humans
Myocardial Infarction* / diagnosis
Uncertainty