2018/11/10 by Sawon Pratiher, Pratiher, Sawon, Subhankar Chattoraj +3
Neuroscience · #Computer Vision and Pattern Recognition (cs.CV) #EEG and Brain-Computer Interfaces #FOS: Biological sciences #FOS: Computer and information sciences #FOS: Electrical engineering #Functional Brain Connectivity Studies #Neural dynamics and brain function #Neurons and Cognition (q-bio.NC) #Signal Processing (eess.SP) #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.1811.04230
openalex publication_date 2018/11/10 · openalex created_date 2022/08/02 · openalex updated_date 2026/07/28
A novel non-stationarity visualization tool known as StationPlot is developed\nfor deciphering the chaotic behavior of a dynamical time series. A family of\nanalytic measures enumerating geometrical aspects of the non-stationarity &\ndegree of variability is formulated by convex hull geometry (CHG) on\nStationPlot. In the Euclidean space, both trend-stationary (TS) &\ndifference-stationary (DS) perturbations are comprehended by the asymmetric\nstructure of StationPlot's region of interest (ROI). The proposed method is\nexperimentally validated using EEG signals, where it comprehend the relative\ntemporal evolution of neural dynamics & its non-stationary morphology, thereby\nexemplifying its diagnostic competence for seizure activity (SA) detection.\nExperimental results & analysis-of-Variance (ANOVA) on the extracted CHG\nfeatures demonstrates better classification performances as compared to the\nexisting shallow feature based state-of-the-art & validates its efficacy as\ngeometry-rich discriminative descriptors for signal processing applications.\n