A real-time ECG QRS detection ASIC based on wavelet multiscale analysis

Abstract

This paper presents for the first time Electrocardiograph (ECG) QRS detection algorithm implemented in Application Specific Integrated Circuit (ASIC). The algorithm based on the dyadic wavelet transform (DYWT) multiscale-product scheme is designed especially for real-time biomedical signal processing applications. The algorithm is evaluated based on the MIT-BIH database and achieves a sensitivity of 99.63% and a positive predictivity of 99.89% of R-waves detection. The results show that the algorithm outperforms several well-known QRS complex detection algorithms. The proposed algorithm is then implemented in ASIC and fabricated with 0.18-μm CMOS technology and it consumes 176 μW at an operating frequency of 1 MHz with a supply voltage of 1.8 V.