Key result
Frequency-filtered ECG spectrograms enable CNN detection of anteroseptal MI with ~99% accuracy.
Why the study?
Traditional manual interpretation of complex ECG signals requires significant effort, knowledge, and time, while existing neural network approaches often demand high computational power and memory.
Does a CNN using STFT-based spectrograms improve binary classification accuracy for anteroseptal myocardial infarction compared to raw ECG signals?
Does a CNN using STFT-based spectrograms improve binary classification accuracy for anteroseptal myocardial infarction compared to raw ECG signals?
Converting ECG signals to denoised and frequency-filtered spectrograms yields highly accurate machine learning detection of anteroseptal myocardial infarction while reducing computational memory usage.
May enhance automated ECG analysis efficiency; extends CNN methods to spectrograms but leaves open prospective clinical validation.
The non-invasive electrocardiogram (ECG) signals are useful in heart condition assessment and are found helpful in diagnosing cardiac diseases. However, traditional ways, i.e., a medical consultation required effort, knowledge, and time to interpret the ECG signals due to the large amount of data and complexity. Neural networks have been shown to be efficient recently in interpreting the biomedical signals including ECG and EEG. The novelty of the proposed work is using spectrograms instead of raw signals. Spectrograms could be easily reduced by eliminating frequencies with no ECG information. Moreover, spectrogram calculation is time-efficient through short-time Fourier transformation (STFT) which allowed to present reduced data with well-distinguishable form to convolutional neural network (CNN). The data reduction was performed through frequency filtration by taking a specific cutoff value. These steps makes architecture of the CNN model simple which showed high accuracy. The proposed approach reduced memory usage and computational power through not using complex CNN models. A large publicly available PTB-XL dataset was utilized, and two datasets were prepared, i.e., spectrograms and raw signals for binary classification. The highest accuracy of 99.06% was achieved by the proposed approach, which reflects spectrograms are better than the raw signals for ECG classification. Further, up- and down-sampling of the signals were also performed at various sampling rates and accuracies were attained.
No takes yet. Share an insight, caveat, or question.
Safdar et al. (2022) studied Anteroseptal Myocardial Infarction (n=18,885). CNN using denoised, frequency-filtered STFT spectrograms vs. CNN using denoised raw ECG signals was evaluated on Binary classification accuracy (Normal vs ASMI). A convolutional neural network using denoised and frequency-filtered ECG spectrograms achieved 99.06% accuracy in classifying anteroseptal myocardial infarction compared to raw signals.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: