Fig. 1
From: Parkinson’s disease tremor prediction using EEG data analysis-A preliminary and feasibility study
Block diagram of the proposed methodology. Accelerometer raw data (panel A, upper), filtered accelerometer data (panel A, middle) and its time-frequency profile (panel A, lower). Multiplication of time-frequency profile of x, y and z-axis accelerometer data (B). According to a thresholding strategy, pre-tremor, tremor and non-tremor (control) segments were detected (C). Based on the onset time of tremor, pre-tremor and non-tremor segments, related EEG segments were extracted by windowing (D). Preprocessing and analyses blocks (E)