What is the clinical evidence from this study?

Population: ECG noise (n=48). Intervention: Feed forward neural network denoising method vs. Noisy ECG signals (no denoising). Primary outcome: Reconstruction Error (RMSE denoised / RMSE noisy) and QRS detection performance.

December 20, 2012Open Access

A Neural Network Approach to ECG Denoising

View Full Paper

Ask AI

Bookmark

Key result

A feed forward neural network denoising method reduced reconstruction error and improved the positive predictivity of QRS detectors on noisy ECG signals.

Population

ECG records from Physionet MIT-BIH Arrhythmia Database with added electrode motion artifact noise

Authors

RRRui RodriguesUniversity of Padua PCPaula CoutoUniversidade Nova de Lisboa

Discussion

Member takes

Implication

May aid automated ECG analysis in noisy settings; leaves open clinical validation and outcome impact.

Structured PICO

Population

48 ECG records from the MIT-BIH Arrhythmia Database used to evaluate a neural network denoising method.

Intervention

ECG denoising method based on a feed forward neural network with three hidden layers

Outcome

Performance of publicly available ECG analysis programs on noisy ECG signals

A feed forward neural network approach can effectively denoise highly corrupted ECG signals, improving the performance of automated analysis programs.

Limitations

For low noise (SNR above 12db), QRS detectors show a slight reduction in sensitivity after applying the denoising method.
The method requires the target channel to be free from noise for some minutes in order to train the neural network.

Cite This Study

Rodrigues et al. (2012) studied ECG noise (n=48). Feed forward neural network denoising method vs. Noisy ECG signals (no denoising) was evaluated on Reconstruction Error (RMSE denoised / RMSE noisy) and QRS detection performance. A feed forward neural network denoising method reduced reconstruction error and improved the positive predictivity of QRS detectors on noisy ECG signals.

synapsesocial.com/papers/6a638b174c0a08ff99b6de28 https://doi.org/10.48550/arxiv.1212.5217

View Full Paper

Ask AI

Bookmark

A Neural Network Approach to ECG Denoising

Discussion

Member takes

Implication

Structured PICO

Limitations

Cite This Study

A Neural Network Approach to ECG Denoising

Discussion

Member takes

Implication

Structured PICO

Limitations

Cite This Study

Also Consider

Also Consider