Design and Analysis of Closed-Loop Decoder Adaptation Algorithms for Brain-Machine Interfaces

Abstract

Closed-loop decoder adaptation (CLDA) is an emerging paradigm for achieving rapid performance improvements in online brain-machine interface (BMI) operation. Designing an effective CLDA algorithm requires making multiple important decisions, including choosing the timescale of adaptation, selecting which decoder parameters to adapt, crafting the corresponding update rules, and designing CLDA parameters. These design choices, combined with the specific settings of CLDA parameters, will directly affect the algorithm's ability to make decoder parameters converge to values that optimize performance. In this article, we present a general framework for the design and analysis of CLDA algorithms and support our results with experimental data of two monkeys performing a BMI task. First, we analyze and compare existing CLDA algorithms to highlight the importance of four critical design elements: the adaptation timescale, selective parameter adaptation, smooth decoder updates, and intuitive CLDA parameters. Second, we introduce mathematical convergence analysis using measures such as mean-squared error and KL divergence as a useful paradigm for evaluating the convergence properties of a prototype CLDA algorithm before experimental testing. By applying these measures to an existing CLDA algorithm, we demonstrate that our convergence analysis is an effective analytical tool that can ultimately inform and improve the design of CLDA algorithms.

Personalized Neuroprosthetics

“…This inherent information uncertainty and daily recording instability often demand recalibration of BMI control algorithms on a daily basis, which may not be conducive to a normal life-style. Although procedures have been developed to seek long-term stability (35), such motor control complexity requires contextual and individualized adjustments to decoding solutions. New statistical frameworks have been proposed to explore the high-dimensionality of neural activity across environments and context, but only recently has implantable wireless neurotechnology been developed to enable collection of neural data for such analysis (36,37).…”

Closed-Loop Decoder Adaptation Shapes Neural Plasticity for Skillful Neuroprosthetic Control

“…Much as neural adaptation has proven beneficial, recent work shows the potential promise of adaptive decoders to improve performance. Closed-loop decoder adaptation (CLDA)-modification of decoder parameters based on closed-loop performance (Dangi et al, 2013)-can reliably improve performance (Taylor et al, 2002;Li et al, 2011;Gilja et al, 2012;Orsborn et al, 2012;Jarosiewicz et al, 2013). CLDA may be particularly useful for compensating for nonstationary neural recordings (Li et al, 2011) and has been shown to produce high-performance BMI control for many months independent of stationary neural recordings (Gilja et al, 2012).…”

Neural shaping with joint optimization of controller and plant under restricted dynamics

“…Neural decoding methods called 'turntaking' alternate between user learning and decoder adaptation [17], [18], but this iterative approach likely settles on a Pareto-optimal controllerplant which is not globally optimal [19]. For example, the closed-loop decoder adaptation (CLDA) method [17], [18] converges to decoder parameters that can induce pathological behavior in cursor movement, like tremor and curling force fields [20].…”

“…For example, the closed-loop decoder adaptation (CLDA) method [17], [18] converges to decoder parameters that can induce pathological behavior in cursor movement, like tremor and curling force fields [20]. Importantly, the authors of that work corrected this pathological behavior by adjusting the related CLDA parameters based on analysis of closed-loop dynamics assuming some control model of the user [20].…”