What does this research mean for the field?

A deep learning model based on one-dimensional convolutional neural networks using discrete Fourier transform features in raw ECG signals can predict the onset of sudden cardiac arrest with 96% accuracy approximately 90 minutes before occurrence. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

December 25, 2023Open Access

1-dimensional convolutional neural networks for predicting sudden cardiac

Structured PICO

Does a 1-dimensional convolutional neural network using discrete Fourier transform of ECG signals accurately predict sudden cardiac death?

Population

38 patients (18 normal patients and 20 sudden cardiac death [SCD] patients) from the MIT-BIH arrhythmia database

Intervention

1-dimensional convolutional neural network (CNN) using discrete Fourier transform (DFT) features (1024 frequency bins) of 1-minute raw electrocardiogram (ECG) segments

Comparator

Other feature extraction methods including raw ECG signal, discrete Wavelet transform (DWT), and time-frequency spectrogram

Outcome

Prediction accuracy of sudden cardiac death (SCD) onset up to 90 minutes before occurrence

A 1D convolutional neural network using discrete Fourier transform of raw ECG signals can accurately predict sudden cardiac death up to 90 minutes before onset, potentially allowing for early life-saving interventions.

Abstract

Sudden cardiac arrest (SCA) is a serious heart problem that occurs without symptoms or warning. SCA causes high mortality. Therefore, it is important to estimate the incidence of SCA. Current methods for predicting ventricular fibrillation (VF) episodes require monitoring patients over time, resulting in no complications. New technologies, especially machine learning, are gaining popularity due to the benefits they provide. However, most existing systems rely on manual processes, which can lead to inefficiencies in disseminating patient information. On the other hand, existing deep learning methods rely on large data sets that are not publicly available. In this study, we propose a deep learning method based on one-dimensional convolutional neural networks to learn to use discrete fourier transform (DFT) features in raw electrocardiogram (ECG) signals. The results showed that our method was able to accurately predict the onset of SCA with an accuracy of 96% approximately 90 minutes before it occurred. Predictions can save many lives. That is, optimized deep learning models can outperform manual models in analyzing long-term signals.

AI에게 질문

Bookmark

View Full Paper