Latent Variate Factorial Principal Component Analysis of Microelectrode Recording of Subthalamic Nuclei Neural Signals with Deep Brain Stimulator in Parkinson Disease

Abstract

Parkinson’s disease (PD) is a complex chronic disorder characterized by four classes of cardinal motor symptom features—tremor, bradykinesia, rigidity, and postural instability. Even though clinical benefits of deep-brain-stimulator(DBS) in subthalamic-nuclei (STN) neurons have been established, albeit, how its mechanisms enhances motor-features reduces tremor and restores and increases motor function have not been fully customized. Also, its objective methods for quantifying efficacy of DBS are lacking. We present a latent variate factorial (or factor) principal component analysis-based method to predict UPDR score objectively. Twelve PD subjects are included in this study. Our hypothesis in this study is that whether the DBS saves the STN neurons and restores motor function after the reduction of tremor or damages. In our long study, the high-frequency stimulation in diseased brain did not damage subthalamic nuclei (STN) neurons but protected. Further, it is risk-free to stimulate STN much prior than it was accepted far so. The latent variate factorial is a statistical mathematical technique principal component analysis (PCA)-based method for computing the effects of DBS in PD. We extracted and then extrapolated microelectrode signal recordings (MER) of STN neurons features. The signal parameters were transformed into a lower-dimensional feature space. So we could predict the disease at an early stage.