AECA: An ambiguous-entropy clustering algorithm for the analysis of resting-state fMRISs of human brain and their functional connections

The ambiguous set theory has recently been introduced to characterize ambiguities inherent in information. Based on ambiguous set and ambiguous entropy (AE), this study presents a clustering algorithm called ambiguous-entropy clustering algorithm (AECA). The proposed AECA applies to analyze the resting-state fMRI signals (RSfMRISs). For the experimental purpose, this study considers RSfMRISs data obtained from resting-state of human brain. Three different data clusters related to RSfMRIS are identified and their statistical significance is analyzed using AECA. We perform the autocorrelation analysis for the clustered RSfMRISs. We test the statistical significance of the results using the Ljung-Box test. The results show that AECA is able to identify the activation of RSfMRISs despite the resting state of the brain. To discover the activation patterns of neurons in the resting state, functional connections in RSfMRISs are investigated. For this purpose, an algorithm called ambiguous-entropy functional connection algorithm (AEFCA) is presented in this study. Using this algorithm, a weighted graph is constructed to show the functional connections. Such a graph is called functional connection network graph (FCNG).