The architecture of the ESP is shown in Fig. 1. The architecture includes the modules of the three stages along with a main FSM that controls the flow of the data between the different stages. The processing of the data is done using fixed point representation.
- Minimum number of intervals only checked.
- More area and More power dissipation.
In Existing system finds only RR, PQ, QP, RT, TR, PS, and SP intervals only. In proposed system is finding others interval between the P, Q, R, S, T peaks and record the data in the memory.
Intervals possibilities: ( Forward and Backward)
- RR, PT, TP, QT, TQ, ST, TS, RP, PR, TR, RT, QR, RQ, SR, RS
ECG Preprocessing Stage
1) ECG Filtering: The block diagram of the preprocessing stage is shown in Fig. 3. Bandpass filtering of the raw ECG signal is the first step in which the filter isolates the predominant QRS energy centered at 10 Hz, and attenuates the low frequencies characteristic of the P and T waves, baseline drift, and higher frequencies associated with electromyographic noise and power line interference.
2) QRS Detection: To detect the QRS complex, the PAT method was used. The PAT is a widely used method, which is based on the amplitude threshold technique exploiting the fact that R peaks have higher amplitudes compared with other ECG wave peaks. With proper filtering of the signal, the method is highly capable of detecting the R peaks in every heartbeat using two threshold levels.
3) T and P Wave Delineation: The delineation of T and P waves is based on a novel technique proposed. The method is based on adaptive search windows along with adaptive thresholds to accurately distinguish T and P peaks from noise peak. In each heartbeat, the QRS complex is used as a reference for the detection of T and P waves in which two regions are demarcated with respect to R peaks.
Feature Extraction Stage
The two main parameters that must be considered while developing a detection (or prediction) system are the complexity and the accuracy of the feature extraction technique in providing the best results. The result of such analysis yielded in a unique set of ECG features, which were found to be the most indicative characteristics of ventricular arrhythmia with a simple-torealize system and high prediction accuracy. The features include interval between P, Q, R, S, T peak values. Fig. 6 shows these intervals on ECG record. It is worth mentioning that the features are extracted from two consecutive heartbeats, unlike other methods that process each heartbeat independently.
The choice of classifier in this paper was the naive Bayes. The naive Bayes classifier is easy to build with no complicated iterative parameter estimation, which makes it particularly useful for hardware implementation. It assumes naive and strong independent distributions between the feature vectors, and this assumption was met, since all the extracted ECG features were independently analyzed and assessed from the beginning. The architecture of the classifier is implemented, as shown in Fig. 7.
Accuracy is increased so the possibility of the detection is increased and low energy and area architecture is designed.