Abstract The significance of Electrocardiogram (ECG) in determining the electrical mobility of human heart is extremely high in the medical world since it assists in visualizing the anomalies of heart and diagnosing the Cardiovascular Diseases (CVDs) in the early stages to prevent the disease complications by providing timely medical interventions without any complexities. The present study introduces an effective signal processing approach for efficiently assessing the ECG signals, in which the Adaptive Median Filter is employed for denoising the input signals without damaging the edge informations and the Hilbert Transform (HT) is implemented for segmenting the noise free signals into multiple regions to improve the feasibility of disease diagnosis in an optimal manner. For extracting the optimal features of the segmented signals, a Crow Search Algorithm (CSA) based approach is employed, which involves in maximizing reliability of detecting the existence of CVD in a wider range whereas the Probabilistic Neural Network (PNN) is used for classifying the extracted features to distinguish the types of cardiac diseases with maximum accuracy. The performance of the overall methodology is evaluated by implementing MATLAB simulink in an efficient manner. Eventually, a comparative analysis is carried out between different classifiers and the obtained outcomes have proved that the Proposed Classifier delivers optimal results with maximum accuracy of 98.9%, which is comparatively better than the other existing classifiers.
Alan : Mühendislik
Dergi Türü : Uluslararası
Benzer Makaleler | Yazar | # |
---|
Makale | Yazar | # |
---|