False alarm reduction for the cardiac arrythmias: AI based comparative analysis

In this paper various classification techniques have been discussed for the comparative analysis of cardiac arrhythmias. Cardiac arrhythmias can be acknowledged as nonuniform variations in heart health. 86% of false alarm ratio has been observed by in intensive care units (ICUs) by WHO and PHYSIONET. High false alarm ratio in intensive care units (ICUs) may be the root cause of a life threat to the patients. Missed ECG (electrocardiography) signals and missed heart rate variability(HRV) indications may cause sudden death. Less accuracy and precision of the monitoring equipment in intensive care units (ICUs) due to the wrong configurations, connections, calibrations originate high false alarm rate that may arise mental depression in patients and death toll. ECG (electrocardiogram) and ABP (arterial blood pressure) are commonly measured to identify the cardiac variations which may differ due to the uneven diagram. Diverse methodologies have been studied for the false alarm reduction in intensive care units (ICUs). Comparative analysis has been done for the fuzzy logic, ANN, Support vector machine (SVM), MLP, MLP-PSO, k-means, fuzzy k-means clustering, fuzzy logic controller (FLC), DWT (discrete wavelet transform) and DWT with ANN. Detailed contrastive survey of diversified schemes depicted that cuckoo-based optimization and Modified cuckoo search (MCS) could perform better. An adequate MCS algorithm is strongly suggested for the false alarm reduction in intensive care units (ICUs) vigorously