Key result
Novel FrFT-Decision Tree ECG model detects cardiac abnormalities with ~99% accuracy.
Why the study?
ECG signals are often affected by artifacts that complicate automatic analysis, especially during feature extraction and beat classification.
A novel ECG signal analysis technique combining Fractional Fourier Transform and a Decision Tree classifier demonstrated high accuracy (98.66%) in detecting cardiac abnormalities.
Electrocardiogram (ECG) signals are widely used for the early diagnosis of cardiac disorders. However, in real-world recordings, ECG signals are often affected by artifacts such as baseline wander, muscle noise, motion artifacts, and power-line interference. These artifacts make it difficult to analyze ECG signals automatically, especially during feature extraction and beat classification. This paper presents a simple ECG signal analysis technique by combining the Fractional Fourier Transform (FrFT) with a Decision Tree (DT) classifier. The FrFT is used to transform the signal into the fractional time-frequency domain, which improves the visibility of important ECG waveform components such as the P-wave, QRS complex, and T-wave. Based on this fractional-domain representation, a set of features is extracted and used as input to the DT classifier. The method was tested on both newly acquired ECG signals and benchmark ECG databases. The performance metrics showed strong results, with recall (Re) of 99.22%, precision (Pr) of 99.07%, and classification accuracy (CA) of 98.66%.
No takes yet. Share an insight, caveat, or question.
Gupta et al. (2026) studied Cardiac abnormalities. Fractional Fourier Transform (FrFT) combined with a Decision Tree (DT) classifier was evaluated on Classification accuracy, recall, and precision. An ECG signal analysis technique combining the Fractional Fourier Transform with a Decision Tree classifier detected cardiac abnormalities with 98.66% accuracy, 99.22% recall, and 99.07% precision.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: