Method for attenuating noise in electrocardiographic signals using the Wavelet transform

• Autores: Garay Porras, Francisco Fernando, Diaz Aliaga, Julio Teodosio, Piscoya Silva, Ulises Abdón, Echevarria Duran, Patrick Fabrizio
• Formato: artículo
• Fecha de Publicación: 2025
• Institución: Universidad Nacional de Ingeniería
• Repositorio: Revistas - Universidad Nacional de Ingeniería
• Lenguaje: español
• OAI Identifier: oai:oai:revistas.uni.edu.pe:article/2322
• Enlace del recurso: https://revistas.uni.edu.pe/index.php/tecnia/article/view/2322
• Nivel de acceso: acceso abierto
• Materia: Trasformada wavelet estacionaria, histograma, curtosis, desviación estándar, distribución normal, valor umbral, niveles de aproximación y detalle; Stationary wavelet transform, histogram, kurtosis, standard deviation, normal distribution, threshold value, approximation levels, levels of detail

This paper presents a noise reduction method for electrocardiographic (ECG) signals with real-time application. The method was evaluated using digitized signals with resolutions of 14, 16, and 24 bits, durations ranging from 30 to 60 seconds, and a sampling frequency of 1000 Hz. Prior to its application, power line interference (220V/60Hz) is attenuated if present. The approach is based on signal decomposition via the Stationary Wavelet Transform (SWT), analyzing 10 detail levels and 10 approximation levels. It was identified that noise is primarily concentrated in the first four levels, occasionally extending to the fifth. To differentiate between useful signal and noise, histograms of the detail coefficients were analyzed, and key metrics such as kurtosis, relative wavelet energy (RWE), and standard deviation were computed to establish adaptive thresholds for effective noise removal. The proposed method achieves significant noise reduction while preserving clinically relevant information. Quantitative evaluation demonstrated an improvement in the signal-to-noise ratio (SNR), validating its effectiveness. Furthermore, its computational efficiency enables implementation in real-time processing systems and integration into high-end microcontrollers, with applications in the diagnosis and monitoring of cardiovascular diseases