Time series real time naive bayes electrocardiogram signal classification for efficient disease prediction using fuzzy rules

Towards the problem of ECG classification and disease prediction, various approaches are analyzed and discussed. However, the methods suffer to achieve higher performance in classification or disease prediction. To improve the performance, an efficient time series real time Naive Bayes ECG classification and disease prediction approach using fuzzy rule is presented in this paper. The method reads the ECG signals available and performs noise removal initially. From the graphs available, the features mentioned above are extracted and if there exist any incomplete or missing signal then the ECG sample has been removed from the data set. Once the preprocessing and feature extraction are done, then the features extracted. With the learned features, the method generates fuzzy rule for different disease class. The proposed algorithm computes posterior probability according to the mapping of different features of fuzzy rule. The classification or disease prediction is performed by measuring multi-feature signal similarity (MFSS). Estimated MFFS value has been used to measure the cardiac disease prone weight (CDPW) towards various classes available. According to the value of CDPW has been used to perform classification or disease prediction.

Automated summary

This paper proposes an efficient method for ECG classification and disease prediction using fuzzy rules, which generates posterior probabilities for different disease classes based on learned features. Classification or disease prediction is then performed using the Multi-Feature Signal Similarity method.