Tracking Neural Modulation Depth by Dual Sequential Monte Carlo Estimation on Point Processes for Brain–Machine Interfaces

Abstract

Classic brain-machine interface (BMI) approaches decode neural signals from the brain responsible for achieving specific motor movements, which subsequently command prosthetic devices. Brain activities adaptively change during the control of the neuroprosthesis in BMIs, where the alteration of the preferred direction and the modulation of the gain depth are observed. The static neural tuning models have been limited by fixed codes, resulting in a decay of decoding performance over the course of the movement and subsequent instability in motor performance. To achieve stable performance, we propose a dual sequential Monte Carlo adaptive point process method, which models and decodes the gradually changing modulation depth of individual neuron over the course of a movement. We use multichannel neural spike trains from the primary motor cortex of a monkey trained to perform a target pursuit task using a joystick. Our results show that our computational approach successfully tracks the neural modulation depth over time with better goodness-of-fit than classic static neural tuning models, resulting in smaller errors between the true kinematics and the estimations in both simulated and real data. Our novel decoding approach suggests that the brain may employ such strategies to achieve stable motor output, i.e., plastic neural tuning is a feature of neural systems. BMI users may benefit from this adaptive algorithm to achieve more complex and controlled movement outcomes.

Kernel-Based Relevance Analysis with Enhanced Interpretability for Detection of Brain Activity Patterns

“…Although, various functions have been tested, the Gaussian function is preferred in pattern classification and machine learning applications since it aims at finding an RKHS with universal approximating ability, not to mention its mathematical tractability (Liu et al, 2011 ; Brockmeier et al, 2013 ). For Gaussian kernels, each pairwise similarity distance between samples is computed as follows (Wang et al, 2016 ):…”

Tracking Fast Neural Adaptation by Globally Adaptive Point Process Estimation for Brain-Machine Interface

“…A dual-model structure has been proposed to estimate the tuning parameter and kinematics at the same time [16], [30], [31]. The dual-model structure includes two steps.…”

“…The better tracking of the modulation depth helped the subject monkey to learn the BC of a target-reach task. Wang et al [31] extended the single-SMCPP into a dual-SMCPP (DSMCPP) structure to predict the continuous trajectory of a monkey's hand movement, as well as the nonlinearity in an LNP encoding model. The DSMCPP method successfully tracked the gradual change in nonlinear neural tuning and improved the kinematics decoding performance compared to the static model.…”

“…1(a). Compared with the existing methods to track the slow change of tuning property [31], the main difference is that the proposed method conducts a global search on the tuning parameters. Thus, the new method is more suitable for the BC task, which requires fast exploration of the preferred direction.…”

“…The results are compared with DSMCPP, which searches the neural tuning parameters in the neighborhood of the previous estimation. Comparing to linear and Gaussian assumptions in [23], DSMCPP and GaPP share the same assumptions on nonlinear evolution and non-Gaussian distribution of states [31]. A Kolmogorov-Smirnov statistical test (KS-plot) is used to qualify how well the time variant neural preferred direction is predicted.…”