Investigation of inter-subject variability and development of optimization technique for electrical brain stimulation – simulation study

Abstract

The electric brain stimulation technique refers to a method of regulating brain activity through electrical stimulation, which has been proven effective in various fields such as brain disorders and cognitive enhancement. However, there is a large inter-subject variability in the stimulation effects, and even results opposite to the intended stimulation effects have been observed. To overcome these limitations, research designed electrical brain stimulation simulation systems. Then, the system successfully predicts the results of specific stimulation montage before clinical trial. In this paper, we propose a methodology for suggesting optimized stimulation montage by considering the anatomical characteristics of each individual using the electrical brain stimulation simulation system. First, we investigated the individual characteristics that can affect the electric brain stimulation effects and conducted studies to quantify their effects. By controlling other factors and changing only the degree of brain atrophy, we found that changes in cerebrospinal fluid have a significant influence comparable to the distance between the electrode and the brain. However, this relationship between cerebrospinal fluid and stimulation effects was not observed in simulation study using actual magnetic resonance imaging data get from ADNI data set. In other words, although the volume difference of cerebrospinal fluid exists statistically between two groups, the statistical significance in stimulation effects was not observed by comparing the two groups. This means that although cerebrospinal fluid has a significant impact on stimulation effects, it cannot explain them solely as other factors also exist in a complex manner in clinical settings. However, as cerebrospinal fluid volume was emphasized as a single factor influencing the stimulation effects, the electric brain stimulation optimization methodology was validated with the study which the optimization methodology could offset the influence of individual differences in cerebrospinal fluid. Deterministic and probabilistic methodologies for electric brain stimulation optimization were both confirmed with their respective advantages and disadvantages, and it was found that minimizing inter-subject variability was achieved by using both methodologies simultaneously.