Fuzzy C-Means Clustering with Minkowski and Euclidean Distance for Cerebral Infarction Classification

Abstract

Cerebral infarction is a condition in which the death of neuronal cells, glial cells and blood vessel system is caused by a lack of oxygen and nutrients. The cause of nerve damage is hypoxia, which is a decrease in oxygen pressure in the alveoli which can cause hypoxemia in brain tissue. Cerebral infarction can also be caused by obstruction of blood flow to the brain so that the brain does not get enough oxygen. This situation is called ischemia. The initial stage of ischemic neurons is characterized by the formation of micro vacuolization, which is characterized by cell size that is still normal or slightly reduced, vacuoles occur in the perikaryon area, which can be found in neurons in the hippocampus and cortical 5-15 minutes after hypoxia. The final sign of cell damage due to ischemia is the nucleus which becomes pyknotic and fragmented. To diagnose the presence or absence of cerebral infarction in the brain it is not enough just to use a CT scan, therefore machine learning will also be used to diagnose the presence or absence of cerebral infarction in the brain. For this reason, the authors propose Fuzzy C-Means Clustering with Minkowski and Euclidean Distance as a classification method that has good accuracy, good precision, good memory, and a good F1-score in calcifying patients whose brains experience infarction or not. In this proposed method, Fuzzy C-Means Clustering with Minkowski and Euclidean Distance is a modification of the Fuzzy C-Means Clustering Algorithm. This modification is proposed to increase the detection capacity of Fuzzy C-Means Clustering. The parameterized Minkowski distance metric is adjusted for implementation with FCM with various settings. The experimental results show that this method can improve the results of the FCM grouping with an accuracy of around 88%.