Signal Quality Evaluation and Processing for QRS Detection in ECG based Smart Healthcare Systems

The next-generation wearable electrocardiogram (ECG) equipment requires signal processing with low battery usage in order to transmit signals when harmful rhythms are recognized and to capture signals when anomalous rhythms are discovered. The visual deflections that are frequently visible on an ECG make up the QRS complex. This research suggests a real-time QRS recognition and R point detection system that is extremely accurate and simple. The recommended ECG signal modification eliminates baseline wandering while also enhancing QRS intervals and controlling P and
$T$
waves. In this work, the peaks and valleys of the converted signal were used to calculate the fiducial point for the QR. The R point could then be determined using four QRS waveform templates, and the initial categorization of cardiac rhythms could be completed simultaneously. On two benchmark datasets, the proposed method's effectiveness is shown. Positive prediction (+P) and detected sensitivity (Se) values for QRS are 99.82 and 99.81 percent, respectively, according to the standard. The outcome demonstrates that the method has a low processing complexity and that real-time software may be successfully executed on both an embedded system and a mobile device.