Analysis And Comparison Of Fuzzy Based Purkinje Fibre Cell Synchronization And Coordination Classification using Potassium Channel Parameters With ANN And KNN Classifier

Abstract

Aim: The goal of this study is to compare fuzzy based purkinje fiber cell(PFC) synchronization and coordination categorization
using potassium(K+ ) channel parameters with K Nearest Neighbor (KNN) and Artificial Neural Network (ANN).
Materials and Methods: The PFC Aslanidi model(AM) is used to classify cardiac arrhythmias of K+ channel at different malfunction conditions. In AM, the K+ channel dynamics are well described based on the PFC dataset. ANN and KNN classifiers are used to classify the different PFC classification. Sample size was calculated by keeping threshold 0.05, G Power 80%, confidence interval 95% and enrolment ratio as 1. 20 samples are taken for each analysis to predict the accuracy of the Novel ANN and KNN classifier. Finally, using the Statistical Package for the Social Science (SPSS) programme, the collected dataset(resultant accuracy) can be calculated.
Result: PFC normal Action Potential (AP) having AP Duration (APD) is 290ms and Total Cycle Length (TCL) is 1000ms at K+ extracellular levels as 5.4mM. Whereas in abnormal AP indicates APD is 280ms and TCL is 900ms at K+ extracellular levels as 10.4mM. A 10% decrease in APD with normal PFC AP and 10% drop in TCL is responsible for Tachycardia (fast heartbeat), Then PFC data(Normal, Tachycardia, Bradycardia) are sent into the Novel ANN and KNN classifier algorithms. Using SPSS software, the results reveal that ANN accuracy is higher than KNN. Group 1 KNN has a mean of 63.410, a standard deviation of 4.3533, and a standard error mean of .9734, whereas group 2 ANN has a mean of 66.055, a standard deviation of 1.0329, and a standard error mean of .2310.

Conclusion: Therefore the changes in PFC K+ channel leads to Hypokalemia which results in cardiac arrhythmia of resting membrane potential and the duration of both the action potential, to a higher extent than the normal. The paper concludes that the ANN classifier predicts better classification in identifying the accuracy of potassium channel parameters when compared to the KNN classifier of PFC synchronization and coordination.