Parkinsonian Tremor Signal Decomposition: Segregating Effects of Deep Brain Stimulation and Medication.

Abstract

Analysis of tremor signal is a crucial part in the study of Parkinsonian subject, specially to understand effectiveness of treatment and progression of disease. Aim of this paper is to segregate the effects of Deep Brain Stimulation (DBS) and medicinal components from Parkinson's disease (PD) tremor signal. Tremor signal has multiple effects embedded in a single channel and identifying the hidden components from it is a challenging process. Conventional methods like Empirical Mode Decomposition (EMD) and Ensemble EMD (EEMD) serve the purpose, however, these methods fail with increase in noise in the signal. We propose the usage of Variational Mode Decomposition (VMD) to identify the underlying hidden components in the tremor signal. It decomposes the tremor signal into different source components, which can be identified as medicinal or DBS components. Results show that VMD is more efficient in disintegrating the medicine and DBS component from the single channel tremor signal, compared to standard EMD and EEMD techniques. This study can help in better understanding of PD tremor suppression mechanism.